Hindawi Publishing Corporation

International Journal of Rotating Machinery

Volume 2015, Article ID 656749, 12 pages
http://dx.doi.org/10.1155/2015/656749

Research Article
Design of a PID Controller for a Linearized Magnetic Bearing

Theodore K. Psonis,1 Pantelis G. Nikolakopoulos,2 and Epaminondas Mitronikas1

1
Laboratory of Electromechanical Energy Conversion, Department of Computer and Electrical Engineering, University of Patras,
Rio, 26500 Patras, Greece
2
Machine Design Laboratory, Department of Mechanical and Aeronautics Engineering, University of Patras, Rio, 26500 Patras, Greece

Correspondence should be addressed to Pantelis G. Nikolakopoulos; [email protected]

Received 12 June 2015; Revised 11 September 2015; Accepted 15 September 2015

Academic Editor: Luis San Andres

Copyright © 2015 Theodore K. Psonis et al. This is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly
cited.

This paper presents the study of magnetic bearings regarding a linear model. Initially, the advantages of magnetic bearings are
referenced, in relation to the existing technology. Subsequently, the linearized model of the system is presented and the need for
closed loop and control of the system is clarified. This need leads to further analysis of linear controllers like P, I, D, PI, PD, and
PID. For each of them, the stability of closed loop system is studied, using the characteristic equation of the system and the Routh-
Hurwitz criterion. To this end, the boundary conditions for the existence of the stability of each of them are found and presented.
After finding the controllers’ characteristic parameters which could provide stability to the system, simulation tests with existence
of white noise follow. Finally, the proposed PID controller performance is examined, based on existing laboratory data, and results
concerning the stability of this controller are presented.

1. Introduction the existence of a higher critical speed possible, while at

the same time contributing to stable operation of the motor.
Magnetic bearings are an improved type of bearings which In the early design of compressors and turbines, it was of
promise to replace the conventional bearings technology like great importance to locate the rotor critical speeds so as
ball or oil film lubricated bearings. This new technology offers to insure that a turbo rotor would not be operating at a
many advantages. The absence of contact between the bearing critical speed. This has to do with the stiffness and damping
and the rotor makes it ideal to be used in vacuum applications coefficients that simulate the oil film behavior and further
and sterile clean rooms. Without friction between the shaft affect the dynamic of the rotor-bearing system [1]. A lot of
and the bearing higher speeds can be attained, up to yield work has been done regarding the stiffness and damping
strength of the shaft. Because of the absence of the lubrication coefficients determination, showing the importance of this
seals, the dimensions and limits of strength can be changed, topic [2]. All of these advantages have contributed to the
allowing the use of larger and more stiff shaft. Increasing the further study of magnetic bearings [3–6]. Although the active
stiffness of the shaft offers many advantages and solutions magnetic bearings have a lot of advantages as described
to the problems occurring at high speeds as well as those above, when coming to their dynamics researchers found
coming from vibrations. Problems such as vibration isolation that they are hard to be mastered. Also it is difficult to
and the ability of the system to exceed critical speeds while get their precise mathematical model. Furthermore, the PID
being resistant to external shocks can be improved. Finally, method is less relying on the mathematical model regarding
the most powerful advantage is the lack of mechanical wear. the control issue. So, PID control is used for the control
This has resulted in lower maintenance costs and longer of the magnetic levitation systems. However, because the
life of the shaft. The advantages of this new technology general PID controllers always have only one couple of
are not just limited only on the bearings but are spreading constant parameters, it is difficult some times to meet the
to the motors that could work with these new bearings. need of control performances. That is why nonlinear control
These motors can have smaller shaft length. This will make strategies have been also developed. PID control schemes
2 International Journal of Rotating Machinery

were presented by the authors in [3, 5, 6]. They presented

the advantages of the use of the linear magnetic bearing
models in combination with PID control laws. These studies
have contributed to the further investigation of the magnetic F1
bearing and to the improvement of it. Farmakopoulos et al.
[7] presented PID models regarding a linear active magnetic
bearing and a hybrid magnetic bearing. They also presented
the limits of the stability of the proposed controllers.
A variation on the conventional PID control is a cascaded 𝜃t
PI/PD position control as presented in [8]. The advantages of
the PI/PD control are transparent design, simple realization,
and a higher closed loop damping and stiffness in comparison W
with the conventional PID control.
The application of 𝐻∞ optimal control design to AMBs
began soon after the introduction of 𝐻∞ control theory by
Zames [9] in 1981. An important requirement in most prac-
tical AMB applications is that the stiffness of the controlled Figure 1: A shaft and a pair of electromagnets.
system, when subjected to unknown dynamic disturbance
forces or loads, should not be below a given value for some
specified frequency range. The new theory was applied by has 5 cm diameter and 5 cm length. It is assumed that the
Herzog and Bleuler [10] to synthesize an 𝐻∞ control and mass density of the shaft is 7.9 g/cm3 for sake of convenient
demonstrate its effectiveness in dealing with “worst case” calculations. The number of turns of each pole is 50 and the
disturbances within the specified frequency range. In [11], gap between the shaft and the inner wall of the bearing set is
Fujita et al. designed and experimentally tested an 𝐻∞ 0.5 mm [8, 12, 13].
controller for robust stability in the presence of various Below the basic operational characteristics are calculated
perturbations and uncertainties in plant parameters for an in order to define and to evaluate the PID control schemes for
AMB system with satisfactory results. the described electromechanical system.
A series of papers have been presented regarding the (1) The weight of shaft can be calculated as below:
nonlinear control models and stability issues in [6, 12, 13]. The
5 cm 2 g
Volterra method is described in [14] while in [15] the LPV 𝑚=𝜋⋅( ) ⋅ 5 cm ⋅ 7.9 3 = 0.775 kg, (1)
control models are also discussed. 2 cm
󵄨 󵄨
In this paper, linear controllers in terms of P, I, D, PI, PD, |𝑊| = 𝑚 ⋅ 󵄨󵄨󵄨𝑔󵄨󵄨󵄨 = 7.595 N. (2)
and PID are analyzed and their stability issues of closed loop
system, using the Routh-Hurwitz criterion, are presented. The weight of the rotor has direction that is shown in Figure 1.
Also, the suitability and the applicability of each of the (2) Calculation of bias current (𝐼bias ) is as follows.
proposed controllers, in a one-dimensional active magnetic At the equilibrium point of the shafts inside the bearing
bearing, and its characteristic parameters, are analyzed and the following equation must be valid:
discussed. For better harmonization with the real data, our
system is disturbed by the presence of white noise. This (𝑊 ⋅ 2 ⋅ 𝑔)
𝐹1 − 𝑊 = 0 󳨐⇒ 𝐼bias = √ ,
method is widely accepted in control studies. The noise values ((𝑁2 ⋅ 𝜇0 ⋅ 𝑆) / (2 ⋅ 𝑔))
vary in a range of 0.001–0.01 V. This helps to better evaluate
the linear controllers. A linear control model can be targeted 𝜃𝑡 (3)
𝑆 = 𝑙 ⋅ (𝜋 ⋅ 𝐷 ⋅ ) = 5.45 ⋅ 10−4 m2 ,
on small perturbations round the equilibrium point of the 360
rotor bearing system, that is why it has some deviations of
a nonlinear model. However, many systems operate in linear 𝐼bias = 2.1 A.
regions, so the PID control schemes may have a wide range For better understanding of the system the block diagram
of applications. The goal of this paper is to investigate the (Figure 2) is shown.
control performance of the magnetic bearings under noise From (2) the following expression is derived:
disturbances. This is done using a simplified bearing model.
𝑁2 ⋅ 𝜇0 ⋅ 𝑆
𝐿0 = = 0.0017121679 H, (4)
2⋅𝑔
2. Models and Formulations
where 𝑔 is gap between shaft and bearing inner wall. 𝑁 is
For better understanding and easier analysis, a linearized the number of turns. 𝐿 0 is nominal inductance. 𝐾𝑓𝑐𝑓 is a
model which consists of a small shaft and a pair of elec- force-current factor and N/A is the units of measurement
tromagnets (Figure 1) is used (1D magnetic bearing). The (newton/ampere):
system is at rest and the shaft can be moved between the
2 ⋅ 𝐿 0 ⋅ cos (𝜋/𝑝)
electromagnets. So, the forces produced by electromagnets 𝐾𝑓𝑐𝑓 = ⋅ 𝐼bias N/A. (5)
should counterbalance the weight of the rotor. The shaft 𝑔
International Journal of Rotating Machinery 3

Noise

Slider X KSD X󳰀
++
Transfer function of Scope
Input + Controller
gain − magnetic bearing gain

Figure 2: Block diagram of the system.

with 𝑝 being the number of poles bearing, 𝑊 × 𝑙 the gasket- Table 1: Effect of controllers in the system.
gap magnet shaft, and 𝑘𝑥 the force-displacement factor.
Controller Rising Rise Restoration Permanent
2 ⋅ 𝐿 0 ⋅ cos (𝜋/𝑝) time time error
𝑘𝑥 = ⋅ 𝐼bias 2 N/m. (6) Minor
𝑔2 Proportional Reduction Increase Reduction
change
The transfer function of the magnetic bearing following the Integral Reduction Increase Increase Elimination
Figure 2 can be written as Minor Minor
Derivative Reduction Reduction
change change
𝐾𝑓𝑐𝑓
. (7)
𝑚 ⋅ 𝑠2 − 𝑘𝑥
3.4. Proportional-Differential Controller (PD). It has as a
Neglecting unbalance and gyroscopic forces, (7) can result the reduction of the rise and the reduction of the
be transformed in the Laplace domain, giving the transfer restoration time. But it has a little effect on the rise time and
function: the steady-state error.
𝑋󸀠 (𝑠) 𝐾𝑓𝑐𝑓
= . (8) 3.5. Proportional-Integral (PI). It has as a result the elimina-
𝑋 (𝑠) 𝑚𝑠2 − 𝑘𝑥 tion of the permanent error and reducing the rising time but
The poles of the transfer function that correspond to an greatly increases the restoration time.
unstable system are 𝑆1,2 = ±√𝑘𝑥 /𝑚, and using these poles,
the AMB plant is an unstable system. 3.6. Proportional-Integral-Differential Controller (PID). It has
For this reason a closed loop control scheme is necessary. as a result the elimination of the permanent error and the
rise. Simultaneously it achieves the reduction of rising and
restoration time [16, 17].
3. Considerations of PID Control
The closed loop system can be stable in the presence of a 3.7. Proportional Controller (P). Assuming the existence of
controller. In linear models the most common and effective proportional control, the system is as follows:
controllers are the P, I, D, PI, PD, and PID. Each of these 𝑋 𝐾P ⋅ 𝐾I
controllers has its own advantages. These features will be = . (9)
𝑋󸀠 𝑚 ⋅ 𝑠2 + 𝐾P ⋅ 𝐾I − 𝑘𝑥
mentioned below.
To ensure system stability, the gain of the proportional
3.1. Proportional Controller (P). The controller output is controller must issue the following condition:
proportional to the error, contributes to the reduction of the 𝐾P × 𝐾𝑓𝑐𝑓 − 𝐾𝑥 ≥ 0 󳨐⇒
rise time, and reduces but never eliminates the steady-state
error. 𝐾𝑥
𝐾P ≥ 󳨐⇒ (10)
𝐾𝑓𝑐𝑓
3.2. Integral Controller (I). It has as a result the elimination of
the permanent fault but deteriorates the transient response. 𝐾P ≥ 8875.1.

3.3. Differential Controller (D). It has the effect of increasing 3.8. Integral Controller (I). Assuming the existence of Integral
the stability of the system by reducing the rise and improving control the system is
the transient response. 𝐾𝑓𝑐𝑓 × 𝐾I
𝑋
The influence of each controller type in the individual 󸀠
= 3
. (11)
characteristics of the system is shown in Table 1. 𝑋 𝑚 × 𝑠 − 𝐾𝑥 × 𝑠 + 𝐾𝑓𝑐𝑓 × 𝐾I
4 International Journal of Rotating Machinery

The roots of the denominator are The roots of the denominator are

𝑠1 𝐾𝑓𝑐𝑓 ⋅ 𝐾I 2 𝐾𝑓𝑐𝑓 ⋅ 𝐾P − 𝐾𝑥 3 𝐾𝑓𝑐𝑓 ⋅ 𝐾I 2

𝑠1 = √ −
3
+ √( ) +( )
2⋅𝑚 3⋅𝑚 2⋅𝑚
⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟
2 2 3
𝐾𝑓𝑐𝑓 ⋅ 𝐾I √2 27 ⋅ 𝑚 ⋅ 𝐾𝑓𝑐𝑓 ⋅ 𝐾I − 4 ⋅ 𝐾𝑥 𝐶
= √−
3
+
2⋅𝑚 108 ⋅ 𝑚3
⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟
𝐾𝑓𝑐𝑓 ⋅ 𝐾I 2 𝐾𝑓𝑐𝑓 ⋅ 𝐾P − 𝐾𝑥 3 𝐾𝑓𝑐𝑓 ⋅ 𝐾I 2
+ √−
3
𝐴 − √( ) +( ) , (18)
2⋅𝑚 3⋅𝑚 2⋅𝑚
⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟
𝐷
𝐾𝑓𝑐𝑓 ⋅ 𝐾I √2 27 ⋅ 𝑚 ⋅ 𝐾𝑓𝑐𝑓 2 ⋅ 𝐾I 2 − 4 ⋅ 𝐾𝑥 3 (12)
+ √−
3
− , 1 1
2⋅𝑚 108 ⋅ 𝑚3
⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟⏟ 𝑠2 = − ⋅ (𝐶 + 𝐷) + 𝑖 ⋅ ⋅ √3 ⋅ (𝐶 − 𝐷) ,
2 2
𝐵
1 1
1 1 𝑠3 = − ⋅ (𝐶 + 𝐷) − 𝑖 ⋅ ⋅ √3 ⋅ (𝐶 − 𝐷) .
𝑠2 = − ⋅ (𝐴 + 𝐵) + 𝑖 ⋅ ⋅ √3 ⋅ (𝐴 − 𝐵) , 2 2
2 2
To ensure the stability of the system, the real part of these
1 1
𝑠3 = − ⋅ (𝐴 + 𝐵) − 𝑖 ⋅ ⋅ √3 ⋅ (𝐴 − 𝐵) . roots must satisfy the following condition:
2 2
Re {𝑠1,2,3 } ≤ 0. (19)
To ensure the stability of the system, the real part of these
roots must satisfy the following inequality: The above inequality is not satisfied for any value of 𝐾𝑖 , 𝐾P .
So, a stable proportional-integral controller is not feasible
Re {𝑠1,2,3 } ≤ 0. (13) for the entire system.

The above inequality cannot be satisfied for any value of 𝐾𝑖 . 3.11. Proportional-Differential Controller (PD). Assuming the
So, integral control of the system is not feasible. existence of proportional-differential controller the system is
the following:
3.9. Differential Controller (D). Assuming the existence of
𝑋 (𝐾D ⋅ 𝑠 + 𝐾P ) ⋅ 𝐾𝑓𝑐𝑓
differential controller, the system becomes
󸀠
= 2
. (20)
𝑋 𝑚 ⋅ 𝑠 + 𝐾𝑓𝑐𝑓 ⋅ 𝐾D ⋅ 𝑠 + 𝐾P ⋅ 𝐾𝑓𝑐𝑓 − 𝐾𝑥
𝑋 𝐾D ⋅ 𝐾𝑓𝑐𝑓 ⋅ 𝑠
= . (14) The roots of the denominator are
𝑋󸀠 𝑚 ⋅ 𝑠2 + 𝐾D ⋅ 𝐾𝑓𝑐𝑓 ⋅ 𝑠 − 𝐾𝑥
𝑠1
The roots of the denominator are 2
−𝐾𝑓𝑐𝑓 ⋅ 𝐾D + √ (𝐾𝑓𝑐𝑓 ⋅ 𝐾D ) − 4 ⋅ (𝐾𝑓𝑐𝑓 ⋅ 𝐾P − 𝐾𝑥 ) ⋅ 𝑚
2 = ,
−𝐾D ⋅ 𝐾𝑓𝑐𝑓 + √ (𝐾D ⋅ 𝐾𝑓𝑐𝑓 ) + 4 ⋅ 𝑚 ⋅ 𝐾𝑓𝑐𝑓 2⋅𝑚
(21)
𝑠1 = , 𝑠2
2⋅𝑚
(15)
2
−𝐾D ⋅ 𝐾𝑓𝑐𝑓 − √ (𝐾D ⋅ 𝐾𝑓𝑐𝑓 ) + 4 ⋅ 𝑚 ⋅ 𝐾𝑓𝑐𝑓
2 −𝐾𝑓𝑐𝑓 ⋅ 𝐾D − √ (𝐾𝑓𝑐𝑓 ⋅ 𝐾D ) − 4 ⋅ (𝐾𝑓𝑐𝑓 ⋅ 𝐾P − 𝐾𝑥 ) ⋅ 𝑚
= .
𝑠2 = . 2⋅𝑚
2⋅𝑚
To ensure system stability, the real part of these roots must
To ensure the stability of the system, the real part of these satisfy the following condition:
roots must satisfy the following inequality:
Re {𝑠1,2 } ≤ 0. (22)
Re {𝑠1,2 } ≤ 0. (16)
Case 1. If the root is negative,
The above inequality cannot be satisfied for any value of 𝐾D . 2
So, differential control for our system is not possible. (𝐾D ⋅ 𝐾𝑓𝑐𝑓 ) − 4 ⋅ 𝑚 ⋅ (𝐾P ⋅ 𝐾𝑓𝑐𝑓 − 𝐾𝑥 ) < 0
2 (23)
(𝐾D ⋅ 𝐾𝑓𝑐𝑓 ) − 4 ⋅ 𝑚 ⋅ 𝐾𝑥
3.10. Proportional-Integral Controller (PI). Assuming the 󳨐⇒ > 𝐾P .
existence of proportional-integral controller the system 4 ⋅ 𝑚 ⋅ 𝐾𝑓𝑐𝑓
becomes
With negative discriminant,
𝑋 𝐾𝑓𝑐𝑓 ⋅ (𝐾I + 𝑠 ⋅ 𝐾P )
= . (17) −𝐾D ⋅ 𝐾𝑓𝑐𝑓
𝑋󸀠 𝑚 ⋅ 𝑠3 + (𝐾𝑓𝑐𝑓 ⋅ 𝐾P − 𝐾𝑥 ) ⋅ 𝑠 + 𝐾𝑓𝑐𝑓 ⋅ 𝐾I Re {𝑠1,2 } = 󳨐⇒ 𝐾D ≥ 0. (24)
2⋅𝑚
International Journal of Rotating Machinery 5

Case 2. If the root is not negative, 4. Simulation Results

𝑠1 In this paragraph, the system is simulated in terms of P,
PD, and PID controllers for which control stability of the
2 system can be guaranteed. For better results, the model as
−𝐾𝑓𝑐𝑓 ⋅ 𝐾D + √ (𝐾𝑓𝑐𝑓 ⋅ 𝐾D ) − 4 ⋅ (𝐾𝑓𝑐𝑓 ⋅ 𝐾P − 𝐾𝑥 ) ⋅ 𝑚 (25) well as stability provided by each controller has been tested
=
2⋅𝑚 by introducing white noise. The white noise in the system
simulates external disturbances which can adversely affect
≤ 0,
the system. Its form is ideal and the noise is independent of
𝑠2 frequency as its spectrum has the same constant value for all
frequencies. The simulated system is shown in Figure 3.
2 For the procedure testing of controllers white noise is
−𝐾𝑓𝑐𝑓 ⋅ 𝐾D − √ (𝐾𝑓𝑐𝑓 ⋅ 𝐾D ) − 4 ⋅ (𝐾𝑓𝑐𝑓 ⋅ 𝐾P − 𝐾𝑥 ) ⋅ 𝑚 (26)
=
introduced with two levels of noise, 0.01 volts and 0.001 volts.
2⋅𝑚 From the below experimental data it is shown that
≤ 0. optimal combination is achieved by PID controller. Using the
PID controller a very small displacement of the rotor from its
equilibrium position is accomplished under the influence of
Adding in members to (25) and (26) the following can be
white noise. Simultaneously, the values of the control current
obtained:
remain small at desired levels. These levels ensure that the
−2 ⋅ 𝐾𝑓𝑐𝑓 ⋅ 𝐾D magnetic bearings do not influence the neighboring magnetic
≤ 0 󳨐⇒ 𝐾D ≥ 0. (27) fields. A further advantage of the controller, known from the
2⋅𝑚 theory, is the most direct smoothing achieved in the system
because of the presence of integral controller.
Using (25) the below expression occurs:

𝐾𝑓𝑐𝑓 ⋅ 𝐾D 4.1. P Controller. Initially setting the proportional gain to

8875.1, the results obtained are shown in Figures 4 and 5.
2 In Figures 4 and 5 the position of the rotor is shown. It has
≥ √ (𝐾𝑓𝑐𝑓 ⋅ 𝐾D ) − 4 ⋅ (𝐾𝑓𝑐𝑓 ⋅ 𝐾P − 𝐾𝑥 ) ⋅ 𝑚 󳨐⇒ been observed that values cannot be accepted as they lead to
(28) large control currents and large values of the divergence of the
4 ⋅ (𝐾𝑓𝑐𝑓 ⋅ 𝐾P − 𝐾𝑥 ) ⋅ 𝑚 ≥ 0 󳨐⇒ shaft.
For optimization of the controller, the proportional gain
𝐾𝑥 has been increased to 12000; results for the same noise levels
𝐾P ≥ .
𝐾𝑓𝑐𝑓 are presented in Figures 6 and 7.
Figures 6 and 7 present the position of the shaft of the
3.12. Proportional-Integral-Differential Controller (PID). bearing. It has been observed that values cannot be accepted
Assuming the existence of proportional-integral-differential as they lead to large control currents. These large control
controller the system is the following: current values presented in P controller could possibly create
a fairly forcible magnetic field, which could adversely affect
𝑋 the magnetic field of the machine creating the movement of
𝑋󸀠
the rotor.
(29)
(𝐾D ⋅ 𝑠2 + 𝐾P ⋅ 𝑠 + 𝐾I ) ⋅ 𝐾𝑓𝑐𝑓
= . 4.2. PD Controller. During the same procedure we set the
𝑚 ⋅ 𝑠3 + 𝐾𝑓𝑐𝑓 ⋅ 𝐾D ⋅ 𝑠2 + (𝐾P ⋅ 𝐾𝑓𝑐𝑓 − 𝐾𝑥 ) ⋅ 𝑠 + 𝐾𝑓𝑐𝑓 ⋅ 𝐾I proportional gain to 12.2083 and the differential gain to
1.9875. For these gain values, results are presented in Figures
In the case of marginal stability the poles of the system are 8 and 9.
zero: Figures 8 and 9 correspond as well in the position of
the rotor center with respect to the bearing center. It has
Re {𝑠1,2,3 } = 0. (30) been observed that values cannot be accepted as they lead
to large values of the divergence of the shaft. To increment
Based on this treaty, the stability of the system is proven after the response speed of the controller, the proportional gain
satisfying the following conditions: has been increased to the value of 192.8569 and the relevant
differential gain to a value of 7.8867. For these values the
𝐾𝐼 ≥ 0, following results in Figures 10 and 11 are obtained.
From Figures 10 and 11 results the fact that the values
𝐾𝑥 correspond to the rotor position can be accepted.
𝐾P ≥ , (31)
𝐾𝑓𝑐𝑓
4.3. PID Controller. At the end of the test procedure the
𝐾D ≥ 0. corresponding gains of PID controller were defined which are
6 International Journal of Rotating Machinery

Band-limited white noise

Gain 1
ki X󳰀
X .5917 +− PID(S) − ++ Ksn
m · s2 + −Kx
Constant Slider gain Scope
PID
controller
Transfer function

Figure 3: Simulation system.

White noise (V) ×10−3 Position of the rotor (m)

2 1
1 0.5
0 0
−1 −0.5
−2 −1
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100
Time (s) Time (s)

i control (A)
10
5
0
−5
−10
0 10 20 30 40 50 60 70 80 90 100
Time (s)

Figure 4: Response with P controller with proportional gain of 8875.1 and white noise of 0.01 volts.

White noise (V) ×10−4 Position of the rotor (m)

0.6 4
0.4 2
0.2
0 0
−0.2 −2
−0.4
−4
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100
Time (s) Time (s)
i control (A)
3
2
1
0
−1
−2
−3
0 10 20 30 40 50 60 70 80 90 100
Time (s)

Figure 5: System response with P controller with proportional gain of 8875.1 and white noise of 0.001 volts.

for the proportional gain 22.1426, for the integral gain 23.2457, Sequentially, the obtained results are shown in Figures 14 and
and for the differential gain 283.4574. The results are shown 15.
in Figures 12 and 13.
Also form Figures 12 and 13 the obtained values could 5. Control Investigation Using
not be accepted. In an effort to optimize the controller, the Experimental Data
gains were changed. The corresponding values have occurred:
(i) of the proportional gain as 319.0187, (ii) of the integral After choosing a PID controller that satisfies the stability
gain as 1273.2977, and (iii) of the differential gain as 7.2565. criteria while giving a satisfactory response rate, the system
International Journal of Rotating Machinery 7

White noise (V) ×10−3 Position of the rotor (m)

2 1
1 0.5
0 0
−1 −0.5
−2 −1
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100
Time (s) Time (s)
i control (A)
10
5
0
−5
−10
0 10 20 30 40 50 60 70 80 90 100
Time (s)

Figure 6: System response with P controller with proportional gain of 12000 and white noise of 0.01 volts.

White noise (V) ×10−4 Position of the rotor (m)

0.6 4
0.4
0.2 2
0 0
−0.2 −2
−0.4
−4
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100
Time (s) Time (s)
i control (A)
3
2
1
0
−1
−2
−3
0 10 20 30 40 50 60 70 80 90 100
Time (s)

Figure 7: System response with P controller with proportional gain of 12000 and white noise of 0.001 volts.

White noise (V) Position of the rotor (m)

2 0.1
1 0.05
0 0
−1 −0.05
−2 −0.1
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100
Time (s) Time (s)

i control (A)
2
1
0
−1
−2
0 10 20 30 40 50 60 70 80 90 100
Time (s)

Figure 8: System response with PD controller with proportional gain of 12.2083 and gifferential gain of 1.9875 and white noise of 0.01 volts.

has been tested using experimental noise data [7]. This data has been chosen is shown in Figure 18. This figure shows that
consists of recorded noise in a range of 0.0048 volts. The the rise time and the recovery time are very small (rise time <
PID parameters were found for the proportional gain to be 0.002 sec, recovery time < 0.015 sec); simultaneously steady-
319.0187, for the integral gain to be 1273.2977, and for the state gain is very small. The results corresponding to the above
differential gain to be 7.2565. Several values have been tested PID values are shown in Figures 16 and 17.
for the gains and corresponding curves of reference tracking. In the next step a fastest controller must be found. So, the
The above values of gains have been selected as they give following PID controller was constructed, with the following
faster response with low current. The faster response which reference tracking which as it is shown in Figure 18.
8 International Journal of Rotating Machinery

White noise (V) Position of the rotor (m)

0.03
0.5 0.02
0.01
0 0
−0.01
−0.5 −0.02
−0.03
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100
Time (s) Time (s)
i control (A)
0.5

−0.5
0 10 20 30 40 50 60 70 80 90 100
Time (s)

Figure 9: System response with PD controller with proportional gain of 12.2083 and differential gain of 1.9875 and white noise of 0.001 volts.

White noise (V) Position of the rotor (m)

2 0.1
1 0.05
0 0
−1 −0.05
−2 −0.1
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100
Time (s) Time (s)
i control (A)
2
1
0
−1
−2
0 10 20 30 40 50 60 70 80 90 100
Time (s)

Figure 10: System response with PD controller with proportional gain of 192.8569 and differential gain of 7.8867 and white noise of 0.01 volts.

The proportional gain is 1526.6269, the integral gain is while the rotor center remains low. So, these PID values can
13330.5096, and the differential gain is 15.8724. Using the be considered as suitable.
above PID values, the system response is shown in Figure 19.
It is depicted from Figure 19 that the proposed PID values 5.1. System Stability. The stability of the system can also be
lead to a constant gain and relative small control currents proved by the mathematical study of the transfer function of
the closed loop system with the above controller PID values:

𝑋
𝑋󸀠
(32)
[𝑠2 (𝐾P + 𝑁 ⋅ 𝐾D ) + 𝑠 (𝐾P ⋅ 𝑁 + 𝐾I ) + 𝐾I ⋅ 𝑁] ⋅ 𝐾𝑓𝑐𝑓 ⋅ 𝑘sn
= ,
𝑚 ⋅ 𝑠4 + 𝑁 ⋅ 𝑚 ⋅ 𝑠3 + ((𝐾P + 𝑁 ⋅ 𝐾D ) ⋅ 𝐾𝑓𝑐𝑓 ⋅ 𝑘sn − 𝑘𝑥 ) 𝑠2 + ((𝐾P ⋅ 𝑁 + 𝐾I ) ⋅ 𝐾𝑓𝑐𝑓 ⋅ 𝑘sn − 𝑘𝑥 ⋅ 𝑁) ⋅ 𝑠 + 𝐾𝑓𝑐𝑓 ⋅ 𝐾I ⋅ 𝑁 ⋅ 𝑘sn

where the proportional gain is 1526.6269, the integral gain is 𝑃3 = −1128.92409 + 1177.08749 ∗ 𝑖,
13330.5096, and the differential gain is 15.8724.
𝑃4 = −1128.92409 − 1177.08749 ∗ 𝑖.
The poles of the above transfer function are
(33)
The transfer function which appears in (32) has 4 poles,
𝑃1 = −9.71450, of which 2 are real and 2 are complex. All of them have
negative real part, which means that they are in the left
𝑃2 = −90.13387, half-plane. This is a necessary and sufficient condition for
International Journal of Rotating Machinery 9

White noise (V) Position of the rotor (m)

0.03
0.5 0.02
0.01
0 0
−0.01
−0.5 −0.02
−0.03
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100
Time (s) Time (s)
i control (A)
0.5

−0.5
0 10 20 30 40 50 60 70 80 90 100
Time (s)

Figure 11: System response with PD controller with proportional gain of 192.8569 and differential gain of 7.8867 and white noise of 0.001
volts.

White noise (V) Position of the rotor (m)

1.5 0.05
1
0.5
0 0
−0.5
−1
−1.5 −0.05
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100
Time (s) Time (s)
i control (A)
2
1
0
−1
−2
0 10 20 30 40 50 60 70 80 90 100
Time (s)

Figure 12: System response with PID controller with proportional gain of 22.1426, integral gain of 23.2457, and differential gain of 283.4574
and white noise of 0.01 volts.

White noise (V) Position of the rotor (m)

0.5 0.01
0.005
0 0
−0.005
−0.5 −0.01
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100
Time (s) Time (s)
i control (A)
0.5

−0.5
0 10 20 30 40 50 60 70 80 90 100
Time (s)

Figure 13: System response with PID controller with proportional gain of 22.1426, integral gain of 23.2457, and differential gain of 283.4574
and white noise of 0.001 volts.

system stability. Observing the real poles of the system, it of the system have small ratio of the real to the imaginary
is perceived that the system has quick response as these part. This helps to cancel the oscillation of the system. All
two poles have not an imaginary part (𝑃1 = −9.7145, the above are perceived by observing Figure 19 which is the
𝑃2 = −90.13387). Simultaneously the two complex poles system response with PID controller.
10 International Journal of Rotating Machinery

White noise (V) Position of the rotor (m)

1.5 0.05
1
0.5
0 0
−0.5
−1
−1.5 −0.05
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100
Time (s) Time (s)
i control (A)
2
1
0
−1
−2
0 10 20 30 40 50 60 70 80 90 100
Time (s)

Figure 14: System response with PID controller with proportional gain of 319.0187, integral gain of 1273.2977, and differential gain of 7.2565
and white noise of 0.01 volts.

White noise (V) Position of the rotor (m)

0.5 0.01
0.005
0 0
−0.005
−0.5 −0.01
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100
Time (s) Time (s)
i control (A)
0.5

−0.5
0 10 20 30 40 50 60 70 80 90 100
Time (s)

Figure 15: System response with PID controller with proportional gain of 319.0187, integral gain of 1273.2977, and differential gain of 7.2565
and white noise of 0.001 volts.

6. Conclusion I: Integral
𝐼bias : Current bias
In this paper the control of a linear model of a magnetic 𝐾D : Derivative gain
bearing is presented. Τhe equations of the closed loop systems 𝐾I : Integral gain
with the individual linear controllers have been analyzed and
𝐾𝑓𝑐𝑓 : Force-current factor
the conditions for control stability have been investigated.
𝐾P : Proportional gain
From the above analysis, the control performance of the
magnetic bearing has been studied with a linear model and 𝐾𝑥 : Force-displacement factor (stiffness coefficients)
three types of controllers, P, PD, and PID. Subsequently, 𝑙: Stack length
these three controllers have been simulated and characteristic 𝐿 0: Nominal inductance
results have been derived with the presence of white noise. 𝑚: Mass of shaft
Then, in order to avoid large values in the associated gains, 𝑁: Number of turns
while keeping the system within certain limits, the PID con- P: Proportional
troller has been selected. For this purpose the corresponding 𝑝: Number of pole bearings
gains have been selected in order to satisfy the required PD: Proportional derivative
boundary conditions. Then the specific controller has been PI: Proportional integral
tested through simulation. Finally, the mathematical model PID: Proportional integral derivative
was studied and the stability of the controller was proved.
𝑆: Cross-sectional area of flux path
𝑊: Weight of shaft
Nomenclature 𝑊 × 𝑙: Gasket-gap magnet shaft
D: Derivative 𝜃𝑡 : Angle between shaft and electromagnet (Figure 1),
𝐹1 : Force from electromagnet 𝜃𝑡 = 25 deg
𝑔: Acceleration of gravity 𝜇0 : Permeability of free space.
International Journal of Rotating Machinery 11

White noise (V) ×10−4 Position of the rotor (m)

1
2
0.5 1
0 0
−0.5 −1
−2
−1
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100
Time (s) Time (s)
i control (A)
2
1
0
−1
−2
0 10 20 30 40 50 60 70 80 90 100
Time (s)

Figure 16: System response with PID controller with proportional gain of 319.0187, integral gain of 1273.2977, and differential gain of 7.2565
and white noise of 0.001 volts.

White noise (V) ×10−4 Position of the rotor (m)

5 5
0 0
−5 −5
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100
Time (s) Time (s)

i control (A) ×10−5 Slider gain

1
5 0.5
0 0
−5 −0.5
−1
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100
Time (s) Time (s)

Figure 17: System response to enforced changes of the 𝑥 position. System with PID controller with proportional gain of 319.0187, integral
gain of 1273.2977, and differential gain of 7.2565 and white noise of 0.001 volts.

1.4
1.2
1
Amplitude

0.8
0.6
0.4
0.2
0
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018
Time (s)

Block response
Tuned response

Figure 18: Reference tracking for PID controller.

12 International Journal of Rotating Machinery

White noise (V) ×10−4 Position of the rotor (m)

5
1
0 0
−1
−5
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100
Time (s) Time (s)

i control (A) ×10−5 Slider gain

2 1
1 0.5
0 0
−1 −0.5
−2 −1
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100
Time (s) Time (s)

Figure 19: System response with PID controller with proportional gain of 1526.6269, integral gain 13330.5096, and differential gain of 15.8724
and white noise of 0.01 volts.

Conflict of Interests Symposium on Magnetic Bearings, pp. 343–348, Tokyo, Japan,

July 1990.
The authors declare that there is no conflict of interests [11] M. Fujita, F. Matsumuro, and M. Shimizu, “H∞ robust control
regarding the publication of this paper. design for a magnetic suspension system,” in Proceedings of the
2nd International Symposium on Magnetic Bearings, pp. 349–
356, Tokyo, Japan, July 1990.
References
[12] W. de Boer, Active magnetic bearings: modelling and control of a
[1] E. J. Gunter, “Lund’s contribution to rotor stability: the indis- five degrees of freedom rotor [M.S. thesis], Eindhoven University
pensable and fundamental basis of modern compressor design,” of Technology, Eindhoven, The Netherlands, 1998.
Journal of Vibration and Acoustics, vol. 125, no. 4, pp. 462–470, [13] G. Schweitzer, “Active magnetic bearings—chances and limi-
2003. tations,” in Proceedings of the IFToMM 6th International Con-
[2] R. Tiwari, A. W. Lees, and M. I. Friswell, “Identification of ference on Rotor Dynamics, vol. 1, pp. 1–14, Sydney, Australia,
dynamic bearing parameters: a review,” The Shock and Vibration September-October 2002.
Digest, vol. 36, no. 2, pp. 99–124, 2004. [14] M. Saffarzadeh, A. Delavarkhalafi, and Z. Nikoueinezhad,
[3] G. Schweitzer and E. H. Maslen, Magnetic Bearings Theory, “Numerical method for solving optimal control problem of
Design and Application to Rotating Machinery, Springer, Berlin, stochastic Volterra integral equations using block pulse func-
Germany, 2009. tions,” The Journal of Mathematics and Computer Science, vol.
11, pp. 22–36, 2014.
[4] S.-L. Chen, S.-Y. Hsu, P.-L. Wang, and D. Juang, “A feasibility
study of a bearingless motor based on three-pole active mag- [15] P. Lopes dos Santos, J. A. Ramos, and J. L. Martins de Carvalho,
netic bearing,” International Journal of Automation and Smart “Identification of linear parameter varying systems using an
Technology, vol. 2, no. 2, pp. 133–140, 2012. iterative deterministic-stochastic subspace approach,” in Pro-
ceedings of the European Control Conference, pp. 4867–4873,
[5] L. Dong and S. You, “Adaptive control of an active magnetic IEEE, Kos, Greece, July 2007.
bearing with external disturbance,” ISA Transactions, vol. 53, no.
[16] P. Denver Timothy, V. Brown Gerald, and H. Jansen Ralph,
5, pp. 1410–1419, 2014.
“Estimator based controller for high speed flywheel mag-
[6] A. Chiba, T. Fukao, and O. Ichikawa, Magnetic Bearings netic bearing system,” NASA Internal Report NASA/TM-2002-
and Bearingless Drives, Elsevier, Amsterdam, The Netherlands, 211795, 2002.
2005.
[17] T. Zhang, Control of magnetic bearings in wind turbines [M.S.
[7] M. G. Farmakopoulos, E. K. Loghis, P. G. Nikolakopoulos, thesis], Delft University of Technology, Delft, The Netherlands,
N. I. Xiros, and C. A. Papadopoulos, “Modeling and control 2010.
of the electrical actuation system of an active hydromagnetic
journal bearing (AHJB),” in Proceedings of the International
Mechanical Engineering Congress and Exposition (ASME ’14),
Montreal, Canada, November 2014.
[8] B. Polajžer, J. Ritonja, G. Štumberger, D. Dolinar, and J.-
P. Lecointe, “Decentralized PI/PD position control for active
magnetic bearings,” Electrical Engineering, vol. 89, no. 1, pp. 53–
59, 2006.
[9] G. Zames, “Feedback and optimal sensitivity: model reference
transformations, multiplicative seminorms, and approximate
inverses,” IEEE Transactions on Automatic Control, vol. 26, no.
2, pp. 301–320, 1981.
[10] R. Herzog and H. Bleuler, “Stiff AMB control using an H-
infinity approach,” in Proceedings of the Second International
 International Journal of

Rotating
Machinery

International Journal of
The Scientific
Engineering Distributed
Journal of
Journal of

Hindawi Publishing Corporation

World Journal
Hindawi Publishing Corporation Hindawi Publishing Corporation
Sensors
Hindawi Publishing Corporation
Sensor Networks
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Journal of

Control Science
and Engineering

Advances in
Civil Engineering
Hindawi Publishing Corporation Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Submit your manuscripts at

http://www.hindawi.com

Journal of
Journal of Electrical and Computer
Robotics
Hindawi Publishing Corporation
Engineering
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

VLSI Design
Advances in
OptoElectronics
International Journal of

International Journal of
Modelling &
Simulation
Aerospace
Hindawi Publishing Corporation Volume 2014
Navigation and
Observation
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014
in Engineering
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014
Engineering
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014
Hindawi Publishing Corporation
http://www.hindawi.com
http://www.hindawi.com Volume 2014

International Journal of
International Journal of Antennas and Active and Passive Advances in
Chemical Engineering Propagation Electronic Components Shock and Vibration Acoustics and Vibration
Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014