Optimal Robust Tuning for 1DoF PI/PID

Control Unifying FOPDT/SOPDT Models

Víctor M. Alfaro ∗,∗∗ Ramon Vilanova ∗∗
∗
Departamento de Automática, Escuela de Ingeniería Eléctrica,
Universidad de Costa Rica, San José, 11501-2060 Costa Rica
(e-mail:[email protected]).
∗∗
Departament de Telecomunicació i d’Enginyeria de Sistemes, Escola
d’Enginyeria, Universitat Autònoma de Barcelona, 08193 Bellaterra,
Barcelona, Spain (e-mail: [email protected]).

Abstract: The aim of the paper is to present tuning equations for one-degree-of-freedom (1DoF)
proportional integral (PI) and proportional integral derivative (PID) controllers. These are
based on a performance/robustness trade-off analysis with first- and second-order plus dead-
time models. On the basis of this analysis a tuning method is developed for 1DoF PI and PID
controllers for servo and regulatory control that allows designing closed-loop control systems
with a specified MS robustness that at the same time have the best possible IAE performance.
The control system robustness is adjusted varying only the controller proportional gain.

Keywords: PID controllers, one-degree-of-freedom controllers, servo/regulatory control,

performance/robustness trade-off.

1. INTRODUCTION Model Control (IMC), and direct synthesis techniques

(Martin et al., 1975; Rivera et al., 1986; Alcántara et al.,
As it has been widely reported, proportional integral 2011).
derivative (PID) type controllers are with no doubt, the
controllers most extensively used in the process industry. Due to the constraints imposed by the 1DoF control al-
Their success is mainly due to their simple structure, easier gorithm it is necessary to develop separate tuning rules
to understand by the control engineer than other most for servo and regulatory control. In addition, the control-
advanced control approaches. system design procedure is usually based on the use of low-
order linear models identified at the control system nor-
In industrial process control applications, the set-point mal operation point. Due to the non-linear characteristics
normally remains constant and good load-disturbance re- found in most industrial processes, it is necessary to con-
jection (regulatory control) is required. There are also sider the expected changes in the process characteristics
applications where the set-point following (servo-control) assuming certain relative stability margins, or robustness
is the more important control task. requirements, for the control system.
Although from their commercial introduction in 1940 Therefore, the design of the closed-loop control system
(Babb, 1990) the original three-term PID control algo- with 1DoF PI and PID controllers must consider the
rithm has evolved into the actual four- or five-term two- main operation of the control system (servo-control or
degree-of-freedom (2DoF) PID control algorithms the vast regulatory control ) and the trade-off of two conflicting
majority of the controllers still in use are of one-degree-of- criteria, the time response performance to set-point or
freedom (1DoF) type. load-disturbances, and the robustness to changes in the
controlled process characteristics. If only the system per-
Since Ziegler and Nichols (1942) presented their PID con-
formance is taken into account, by using for example an
troller tuning rules, a great number of other procedures
integrated error criteria (IAE, ITAE or ISE) or a time
have been developed as revealed in O’Dwyer (2006) re-
response characteristic (overshoot, rise-time or settling-
view. Some of them consider only the system performance
time) as in Huang and Jeng (2002), and Tavakoli and
(López et al., 1967; Rovira et al., 1969), its robustness
Tavakoli (2003), the resulting closed-loop control system
(Åström and Hägglund, 1984), or a combination of perfor- probably will have a very low robustness. On the other
mance and robustness (Ho et al., 1999). hand, if the system is designed to have high robustness as
There are tuning rules optimized for regulatory control in Hägglund and Åström (2002) and if the performance
operation (López et al., 1967) or optimized for servo- of the resulting system is not evaluated, the designer will
control operation (Tavakoli and Tavakoli, 2003). There are not have any indication of the cost of having such highly
also authors that present separate sets of rules for each robust system. Control performance and robustness are
operation (Zhuang and Atherton, 1993; Kaya, 2004). For taken into account in Shen (2002), and Tavakoli et al.
the servo-control operation there is an important group (2005) optimizing its IAE or ITAE performance but they
of tuning rules based on zero-pole cancellation, Internal just guarantee the usual minimum level of robustness.
Figure 1. Closed-Loop Control System
To have an indication of the performance loss when the
control system robustness is increased, using MS as a mea- Figure 2. PID Closed-Loop Control System
sure, a performance/robustness analysis was conducted for
1DoF and 2DoF PI and PID control systems with first- Equation (2) may be rearranged, for analysis purposes, as
(FOPDT) and second-order plus dead-time (SOPDT) follows  
models (Alfaro et al., 2010). 1
u(s) =Kp 1 + r(s)
Ti s
Based on this performance/robustness analysis, tuning  
rules are proposed for servo and regulatory 1DoF PI 1 Td s
− Kp 1 + + y(s), (3)
and PID controllers for four MS robustness levels in the Ti s 0.1Td s + 1
range from 1.4 to 2.0, to design robust closed-loop control or in the compact form shown in Fig. 2 as
systems that at the same time have the best possible
u(s) = Cr (s)r(s) − Cy (s)y(s), (4)
performance under the IAE criteria. The presented tuning
rules integrate in a single set of equations the tuning where Cr (s) is the set-point controller transfer function
of controllers for first- and second-order plus dead-time and Cy (s) is the feedback controller transfer function.
process models. The output of the closed-loop control system varies with
The rest of the paper is organized as follows: the transfer a change in any of its the inputs as:
functions of the controlled process model, the controller, Cr (s)P (s) P (s)
and the closed-loop control system are presented in Section y(s) = r(s) + d(s), (5)
1 + Cy (s)P (s) 1 + Cy (s)P (s)
2; the performance/robustness analysis is summarized in or
Section 3; the proposed Optimal and Robust Tuning is y(s) = Myr (s)r(s) + Myd (s)d(s), (6)
presented in Section 4 and particular examples of the
performance/robustness trade-off are shown in Section 5. where Myr (s) is the transfer function from the set-point to
The paper ends with some conclusions. the controlled process variable and is known as the servo
control closed-loop transfer function; Myd (s) is the trans-
fer function from the load disturbance to the controlled
2. PROBLEM FORMULATION process variable and is known as the regulatory control
closed-loop transfer function.
Consider a closed-loop control system, as shown in Fig.
1, where P (s) and C(s) are the controlled process model The performance of the closed-loop control system is
and the controller transfer function, respectively. In this evaluated using the IAE cost functional given by
system, r(s) is the set point; u(s), the controller output
Z ∞ Z ∞
.
signal; d(s), the load disturbance; and y(s), the controlled Je = |e(t)| dt = |y(t) − r(t)| dt. (7)
0 0
process variable.
The controller parameters in the servo-control closed-loop
The controlled process is represented by an SOPDT model transfer function, Myr , are the same than the controller
given by the general transfer function parameters in the regulatory control closed-loop transfer
Ke−Ls L function, Myd . Therefore it is not possible to obtain a
P (s) = , τo = , (1) single set of controller parameters θc that optimize, at
(T s + 1)(aT s + 1) T the same time, the control system response to a set-point
where K is the gain; T , the main time constant; a, the step change and the control system response to a load-
ratio of the two time constants (0 ≤ a ≤ 1.0); L, the dead- disturbance step change.
time; and τo , the normalized dead time. The model transfer
function (1) allows the representation of FOPDT processes The performance (7) is evaluated for a step change in the
(a = 0), over damped SOPDT processes (0 < a < 1), and set-point, Jer and in the load-disturbance, Jed .
dual-pole plus dead-time (DPPDT) processes (a = 1). The peak magnitude of the sensitivity function is used as
The process is controlled with a 1DoF PID controller an indicator of the system robustness (relative stability).
whose output is as follows (Åström and Hägglund, 1995): The maximum sensitivity for the control system is defined
     as
1 Td s 1
u(s) = Kp 1+ e(s) − y(s) , (2) .
MS = max |S(jω)| = max . (8)
Ti s αTd s + 1 ω ω |1 + Cy (jω)P (jω)|
where Kp is the controller proportional gain; Ti , the
integral time constant; Td , the derivative time constant; If the system robustness (8) is not taken into account for
and α, the derivative filter constant. Then the controller the design, the controller parameters may be optimized
parameters to tune are θc = {Kp , Ti , Td }. Usually, α = to maximize the system performance or to achieve the
0.10 (Corripio, 2001). minimum value of the cost functional in (7), using Myr
for set point changes (Jer
o
) and Myd for load disturbance Therefore, the second optimization provided the controller
changes (Jed ).
o parameters θco2 required to formulate a system with the
target robustness (8), MSt , and with the best performance
Because of the control system performance/robustness allowed when using the IAE criteria (7), Jer or Jed .
trade-off, if a robustness constraint is included into the
design then, it is expected that the actual system perfor- The performance/robustness analysis of the resulting in
mance will be reduced (Je ≥ Jeo ). Then, the performance PI and PID closed-loop control systems pointed out the
degradation factor defined as existing trade-off between them. As shown in Alfaro et al.
. Jo (2010), in general performance optimized 1DoF PI con-
Fp = e , Fp ≤ 1, (9) trollers are more robust than the PIDs but their optimal
Je
performance is lower. The performance optimized regula-
is used to evaluate the performance/robustness trade-off.
tory control systems, for both PI and PID, are less robust
3. PERFORMANCE/ROBUSTNESS TRADE-OFF than the servo-control ones, requiring also more perfor-
ANALYSIS mance degradation, lower degraded performance factor, to
reach the same robustness level.
To evaluate the performance degradation when the system
robustness is increased, the following steps, as they were 4. UNIFIED SIMPLE OPTIMAL ROBUST TUNING
presented in Alfaro et al. (2010), were followed. FOR 1DOF PI AND PID CONTROLLERS (U SORT1 )

3.1 1DoF Controllers Optimum Performance One of the purposes of this contribution is try to capture
in a single set of equations the performance/robustness
For the 1DoF servo- and regulatory-control performance- trade-off. This is with no doubt a novel feature as the first-
optimized PI and PID controllers, the parameters θco = and second-order models are considered at once, without
{Kpo , Tio , Tdo } were obtained using the cost functional (7) forcing a distinction with respect to neither the model used
such that nor the controller structure. The other purpose is that
.
Jeo = Je (θco ) = min Je (θc ), (10) these robust tuning equations be as simple as possible.
θc
for (1) with a ∈ {0, 0.25, 0.5, 0.75, 1} and ten τo in the Analysis of the regulatory and servo-control PI and PID
range from 0.05 to 2.0, for set-point and load-disturbance controllers parameters shows that for a model with a
step changes. The robustness of the control systems that given time constants ratio a, increasing the control system
deliver the optimal performance was evaluated by using robustness by decreasing MSt , results in a substantial
MS . reduction in Kp . However, this increase in the robustness
has negligible effect on Ti and Td , except in the case
3.2 1DoF Controllers Degraded Performance of models with a very low τo (when high robustness is
required).
To increase the control-loop robustness, a target perfor-
mance degradation factor, Fpt , was included in the cost On the basis of this observation, equations that are inde-
functional, as follows pendent of the target robustness level can be obtained for
the controller integral time constant and derivative time
. Jeo constant, as follows:
JFp = J(θc , Fpt ) = − Fpt , (11)
Je (θc ) Ti = F(T, τo , a), Td = G(T, τo , a). (16)
for obtaining the PI and PID (servo and regulatory con-
trol) parameters θco1 such that With these equations at hand, the controller proportional
. gains are readjusted to match a target robustness to obtain
JFop = JFp (θco1 , Fpt ) = min JFp (θc , Fpt ). (12)
θc equations given by the following
When Fpt was decreased, the control-system robustness Kp = H(K, τo , a, MSt ). (17)
was increased to the target level, MSt . For FOPDT and SOPDT models with τo in the range from
With starting point as the original unconstrained (from 0.1 to 2.0 and four MSt values the normalized 1DoF PI
the point of view of robustness) optimal parameters θco1 , and PID controller parameters can be obtained using the
a second optimization was conducted using the cost func- process model parameters, θp = {K, T, a, L, τo }, for servo-
tional control and regulatory control from the following relations:
.
JMS = J(θc , MSt ) = MS (θc ) − MSt , (13) • Regulatory control operation:
in order to achieve the target robustness. The robust
controller parameters, θco2 , are such that
. .
o
JM = JMS (θco2 , MSt ) = min JMS (θc , MSt ). (14) κp = Kp K = a0 + a1 τoa2 , (18)
S
θc
. Ti
τi = = b0 + b1 τob2 , (19)
For the analysis, four target robustness levels were consid- T
ered, MSt ∈ {2, 1.8, 1.6, 1.4}. . Td
τd = = c0 + c1 τoc2 , (20)
T
Finally, the performance degradation factor required for
obtaining MSt in (14) was evaluated as follows
Jeo
Fp (MSt ) = . (15)
Je (θco2 )
 Table 1. Regulatory Control PI Tuning
Table 2. Regulatory Control PID Tuning
Controlled process time constants ratio a
0.0 0.25 0.50 0.75 1.0 Controlled process time constants ratio a
Target robustness MSt = 2.0 0.0 0.25 0.50 0.75 1.0
a0 0.265 0.077 0.023 -0.128 -0.244 Target robustness MSt = 2.0
a1 0.603 0.739 0.821 1.035 1.226 a0 0.235 0.435 0.454 0.464 0.488
a2 -0.971 -0.663 -0.625 -0.555 -0.517 a1 0.840 0.551 0.588 0.677 0.767
Target robustness MSt = 1.8 a2 -0.919 -1.123 -1.211 -1.251 -1.273
a0 0.229 0.037 -0.056 -0.160 -0.289 Target robustness MSt = 1.8
a1 0.537 0.684 0.803 0.958 1.151 a0 0.210 0.380 0.400 0.410 0.432
a2 -0.952 -0.626 -0.561 -0.516 -0.472 a1 0.745 0.500 0.526 0.602 0.679
Target robustness MSt = 1.6 a2 -0.919 -1.108 -1.194 -1.234 -1.257
a0 0.175 -0.009 -0.080 -0.247 -0.394 Target robustness MSt = 1.6
a1 0.466 0.612 0.702 0.913 1.112 a0 0.179 0.311 0.325 0.333 0.351
a2 -0.911 -0.578 -0.522 -0.442 -0.397 a1 0.626 0.429 0.456 0.519 0.584
Target robustness MSt = 1.4 a2 -0.921 -1.083 -1.160 -1.193 -1.217
a0 0.016 -0.053 -0.129 -0.292 -0.461 Target robustness MSt = 1.4 †
a1 0.476 0.507 0.600 0.792 0.997 a0 0.155 0.228 0.041 0.231 0.114
a2 -0.708 -0.513 -0.449 -0.368 -0.317 a1 0.455 0.336 0.571 0.418 0.620
a2 -0.939 -1.057 -0.725 -1.136 -0.932
b0 -1.382 0.866 1.674 2.130 2.476
†Valid only for τo ≥ 0.40 if a ≥ 0.25
b1 2.837 0.790 0.268 0.112 0.073
b2 0.211 0.520 1.062 1.654 1.955 b0 -0.198 0.095 0.132 0.235 0.236
b1 1.291 1.165 1.263 1.291 1.424
• Servo-control operation: b2 0.485 0.517 0.496 0.521 0.495
c0 0.004 0.104 0.095 0.074 0.033
. c1 0.389 0.414 0.540 0.647 0.756
κp = Kp K = a0 + a1 τoa2 , (21) c2 0.869 0.758 0.566 0.511 0.452
2
T
. i b 0 + b τ
1 o + b τ
2 o
τi = = , (22)
T b3 + τ o
Table 3. Servo-Control PI Tuning
. Td
τd = = c0 + c1 τoc2 , (23)
T Controlled process time constants ratio a
The value of the constants ai , bi , and ci in (18) to (23) are 0.0 0.25 0.50 0.75 1.0
listed in Tables 1 to 4. As noted in these Tables only the Target robustness MSt = 1.8
ai constants for Kp calculation depend on the robustness a0 0.243 0.094 0.013 -0.075 -0.164
level MS . a1 0.509 0.606 0.703 0.837 0.986
a2 -1.063 -0.706 -0.621 -0.569 -0.531
Equations (18) to (23) provide a direct controller tuning Target robustness MSt = 1.6
for the FOPDT (a = 0) and the DPPDT (a = 1) models. a0 0.209 0.057 -0.010 -0.130 -0.220
In the case of the SOPDT models with a ∈ / {0.25, 0.5, 0.75} a1 0.417 0.528 0.607 0.765 0.903
the set of controller parameters must be obtained by linear a2 -1.064 -0.667 -0.584 -0.506 -0.468
Target robustness MSt = 1.4
interpolation between the two sets of parameters obtained
a0 0.164 0.019 -0.061 -0.161 -0.253
with the adjacent a values used in the optimization.
a1 0.305 0.420 0.509 0.636 0.762
The performance/robustness analysis also shows that the a2 -1.066 -0.617 -0.511 -0.439 -0.397
PI controllers with performance optimized parameters for b0 14.650 0.107 0.309 0.594 0.625
servo-control operation produce control systems with a b1 8.450 1.164 1.362 1.532 1.778
robustness MS ≈ 1.8. Then, the minimum robustness level b2 0.0 0.377 0.359 0.371 0.355
b3 15.740 0.066 0.146 0.237 0.209
of MS = 2.0 is exceeded in this case.
With a maximum absolute deviation from the target
robustness MSt of 4.09% and an average deviation of
only 0.70% the proposed uSORT1 tuning may be consid- 10
Servo (a=0.0)
Servo (a=0.25)
ered as a global robust tuning method with levels MSt ∈ 9
Servo (a=0.50)
Servo (a=0.75)
{2.0, 1.8, 1.6, 1.4} for FOPDT and SOPDT models with 8 Servo (a=1.0)

normalized dead-times in the range from 0.1 to 2.0. 7

Regul (a=0.0)
Regul (a=0.25)
Regul (a=0.50)

Equations (18) to (20) and (21) to (23) were obtained 6 Regul (a=0.75)
d
τ /τ

Regul (a=1.0)

for tuning Standard PID controllers. It is know that an

i

equivalent Serial PID controller only exists if Ti /Td ≥ 4. 4

As can be seen from Fig. 3 for the uSORT1 regulatory 3

control τi /τd < 4, then there is no Serial PID equivalent in 2

this case, and that for the uSORT1 servo-control in general 1

0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
τi /τd ≥ 4 for time constant dominant models (τo ≤ 1.0). τ
o

In the particular case of FOPDT controlled process models

the servo-control Serial PID equivalent exists for τo ≤ 1.4. Figure 3. Servo and Regulatory Control τi /τd Ratio
 Table 4. Servo-Control PID Tuning Table 6. P1 Regulatory Control Operation
Controlled process time constants ratio a uSORT1 MSd MEB
0.0 0.25 0.50 0.75 1.0 2.0 1.8 1.6 1.4 IAE
Target robustness MSt = 2.0 PI Controller
a0 0.377 0.502 0.518 0.533 0.572 Kp 0.885 0.779 0.651 0.500 -
a1 0.727 0.518 0.562 0.653 0.728 Ti 2.576 -
a2 -1.041 -1.194 -1.290 -1.329 -1.363 MSr 2.01 1.81 1.61 1.42 -
Target robustness MSt = 1.8 Jed /∆d 2.910 3.305 3.960 5.156 -
a0 0.335 0.432 0.435 0.439 0.482 PID Controller
a1 0.644 0.476 0.526 0.617 0.671 Kp 1.108 0.984 0.829 0.626 1.293
a2 -1.040 -1.163 -1.239 -1.266 -1.315 Ti 1.867 1.971
Target robustness MSt = 1.6 Td 0.614 0.569
a0 0.282 0.344 0.327 0.306 0.482 MSr 2.02 1.82 1.61 1.40 2.36
a1 0.544 0.423 0.488 0.589 0.622 Jed /∆d 1.969 2.215 2.593 3.303 1.666
a2 -1.038 -1.117 -1.155 -1.154 -1.221
Target robustness MSt = 1.4 Table 7. P2 Servo-Control Operation
a0 0.214 0.234 0.184 0.118 0.147
a1 0.413 0.352 0.423 0.575 0.607 uSORT1 MSd MEB
a2 -1.036 -1.042 -1.011 -0.956 -1.015 2.0 1.8 1.6 1.4 IAE
PI Controller
b0 1687 0.135 0.246 0.327 0.381
Kp - 0.711 0.590 0.441 -
b1 339.2 1.355 1.608 1.896 2.234
Ti - 3.421 -
b2 39.86 0.333 0.273 0.243 0.204
MSr - 1.83 1.62 1.41 -
b3 1299 0.007 0.003 -0.006 -0.015
Jer /∆r - 4.311 4.831 6.469 -
c0 -0.016 0.026 -0.042 -0.086 -0.110 PID Controller
c1 0.333 0.403 0.571 0.684 0.772 Kp 1.110 0.989 0.839 0.625 1.497
c2 0.815 0.613 0.446 0.403 0.372 Ti 4.264 5.121
Td 0.921 0.812
Table 5. P1 Servo-Control Operation MSr 1.98 1.79 1.61 1.40 2.78
Jer /∆r 3.385 3.596 4.234 5.687 3.798
uSORT1 MSd MEB
2.0 1.8 1.6 1.4 IAE Table 8. P2 Regulatory Control Operation
PI Controller
Kp - 0.778 0.646 0.482 - uSORT1 MSd MEB
Ti - 2.546 - 2.0 1.8 1.6 1.4 IAE
MSr - 1.81 1.61 1.40 - PI Controller
Jer /∆r - 2.947 3.282 4.392 - Kp 0.838 0.740 0.613 0.461 -
PID Controller Ti 3.743 -
Kp 1.132 1.003 0.846 0.642 1.174 MSr 2.03 1.83 1.62 1.42 -
Ti 3.022 3.085 Jed /∆d 4.466 5.059 6.102 8.098 -
Td 0.495 0.589 PID Controller
MSr 2.0 1.80 1.60 1.40 2.21 Kp 1.037 0.951 0.801 0.620 1.539
Jer /∆r 2.458 2.512 2.976 3.918 2.481 Ti 2.454 2.971
Td 1.108 0.883
5. EXAMPLES MSr 1.93 1.79 1.60 1.41 2.94
Jed /∆d 2.848 3.094 3.605 4.456 2.141
For comparison of the performance and robustness ob-
tained with the proposed uSORT1 method we use the control operation of P2 are listed in Table 7 and Table 8,
Madhuranthakam et al. (2008) [MEB] tuning rules for respectively.
Standard PID controllers that optimize the IAE criteria From Tables 5 to 8 it is noted that for same robust-
for servo- and regulatory control operation. ness design level (MSd ) the PID controllers deliver more
First, we consider the FOPDT process given by performance than the PI controllers. They also show the
performance/robustness trade-off, an increment in control
1.2e−1.5s system robustness always reduces its performance. For
P1 (s) = .
2s + 1 example, to increase the robustness reducing MSd from 1.8
to 1.6 produces a 11 to 20% reduction in the control system
The controller parameters and the control system per-
performance.
formance and robustness for servo-control and regulatory
control operation of P1 are listed in Table 5 and Table 6, It is also noted that the performance optimized MEB
respectively. control systems have low robustness, MS > 2.0 in all cases.
Although the MEB controllers are performance optimized
As a second model we consider the SOPDT process given
the servo-control uSORT1 PID controllers for MSd = 2.0
by
produce control systems that are more robust and that at
1.2e−1.5s the same time have better performance.
P2 (s) = .
(2s + 1)(s + 1)
The P2 control system responses to a 10% set-point and
The controller parameters and the control system per- load-disturbance step changes are shown in Fig. 4 and Fig.
formance and robustness for servo-control and regulatory 5, respectively.
 85
Åström, K.J. and Hägglund, T. (1984). Automatic tuning
of simple regulators with specification on phase and
80
amplitude margins. Automatica, 20(5), 645–651.
y(t), r(t) (%)

75
Åström, K.J. and Hägglund, T. (1995). PID Controllers:
70 Theory, Design and Tuning. Instrument Society of
65 America, Research Triangle Park, NC, USA.
0 1 2 3 4 5 6 7 8 9 10
Babb, M. (1990). Pneumatic Instruments Gave Birth to
uSORT1 PID MS = 2.0
90
uSORT1 PID MS = 1.6
Automatic Control. Control Engineering, 37(12), 20–22.
85 MEB PID IAE Opt. Corripio, A.B. (2001). Tuning of Industrial Control Sys-
tems. ISA - The Instrumentation, Systems, and Au-
u(t) (%)

80

75 tomation Society, Research Triangle Park, NC, USA.,

70 2nd. edition.
65
0 1 2 3 4 5 6 7 8 9 10 Hägglund, T. and Åström, K.J. (2002). Revisiting the
time
Ziegler-Nichols tuning rules for PI control. Asian Jour-
nal of Control, 4, 354–380.
Figure 4. Model P2 Servo-Control Responses Ho, W.K., Lim, K.L., Hang, C.C., and Ni, L.Y. (1999).
Getting more Phase Margin and Performance out of PID
85 controllers. Automatica, 35, 1579–1585.
Huang, H.P. and Jeng, J.C. (2002). Monitoring and ass-
y(t), d(t) (%)

80
esment of control performance for single loop systems.
75 Ind. Eng. Chem. Res., 41, 1297–1309.
70 Kaya, I. (2004). Tuning PI controllers for stable process
0 2 4 6 8 10 12 14 16 18 20 with specifications on Gain and Phase margings. ISA
uSORT1 PID MS = 2.0 Transactions, 43, 297–304.
75
uSORT1 PID MS = 1.6 López, A.M., Miller, J.A., Smith, C.L., and Murrill, P.W.
70
MEB PID IAE Opt. (1967). Tuning Controllers with Error-Integral Criteria.
u(t) (%)

65
Instrumentation Technology, 14, 57–62.
60
Madhuranthakam, C.R., Elkamel, A., and Budman, H.
55

50
(2008). Optimal tuning of PID controllers for FOPDT,
0 2 4 6 8 10
time
12 14 16 18 20 SOPDT and SOPDT with lead processes. Chemical
Engineering and Processing, 47, 251–264.
Martin, J., Corripio, A.B., and Smith, C.L. (1975). Con-
Figure 5. Model P2 Regulatory Control Responses troller Tuning from Simple Process Models. Instrumen-
6. CONCLUSIONS tation Technology, 22(12), 39–44.
O’Dwyer, A. (2006). Handbook of PI and PID Controller
Based on a performance (IAE ) - robustness (MS ) analysis Tuning Rules. Imperial College Press, London, UK, 2nd
tuning relations are proposed that unifies the treatment of edition.
one-degree-of-freedom (1DoF) PI and PID controllers and Rivera, D.E., Morari, M., and Skogestad, S. (1986). Inter-
the use of first- and second-order plus dead-time (FOPDT, nal Model Control. 4. PID Controller Desing. Ind. Eng.
SOPDT) models for servo- and regulatory control systems. Chem. Des. Dev., 25, 252–265.
Rovira, A., Murrill, P.W., and Smith, C.L. (1969). Tuning
The proposed Unified Simple Optimal and Robust Tuning
Controllers for Setpoint Changes. Instrumentation &
for 1DoF PI/PID controllers (uSORT1 ) allows to adjust
Control Systems, 42, 67–69.
the control system robustness varying only the controller
Shen, J.C. (2002). New tuning method for PID controller.
proportional gain.
ISA Transactions, 41, 473–484.
ACKNOWLEDGMENTS Tavakoli, S., Griffin, I., and Fleming, P.J. (2005). Robust
PI controller for load disturbance rejection and setpoint
This work has received financial support from the Spanish regulation. In IEEE Conference on Control Applica-
CICYT program under grant DPI2010-15230. Also, the tions. Toronto, Canada.
financial support from the University of Costa Rica is Tavakoli, S. and Tavakoli, M. (2003). Optimal tuning of
greatly appreciated. PID controllers for first order plus time delay models
using dimensional alalysis. In The Fourth International
REFERENCES Conference on Control and Automation (ICCA’03).
Montreal, Canada.
Alcántara, S., Zhang, W.D., Pedret, C., Vilanova, R., and
Zhuang, M. and Atherton, D.P. (1993). Automatic tuning
Skogestad, S. (2011). IMC-like analytical hinf design
of optimum PID controllers. IEE Proceedings D, 140(3),
with S/SP mixed sensitivity consideration: Utility in
216–224.
PID tuning guidance. Journal of Process Control, 21,
Ziegler, J.G. and Nichols, N.B. (1942). Optimum settings
554–563.
for Automatic Controllers. ASME Transactions, 64,
Alfaro, V.M., Vilanova, R., Méndez, V., and Lafuente,
759–768.
J. (2010). Performance/Robustness Tradeoff Analysis
of PI/PID Servo and Regulatory Control Systems. In
IEEE International Conference on Industrial Technol-
ogy (ICIT 2010). 14-17 March, Viña del Mar, Chile.