Elsevier

Mechatronics

Volume 31, October 2015, Pages 222-233
Mechatronics

Analytical design of fractional order PID controllers based on the fractional set-point weighted structure: Case study in twin rotor helicopter

https://doi.org/10.1016/j.mechatronics.2015.08.008Get rights and content

Highlights

  • Presenting a new analytical method for designing set-point weighted fractional order PID controller.

  • Presenting a simple analytical method for designing filtered fractional order PI controller.

  • Verifying the theoretical results on a laboratory scale twin rotor helicopter experimentally.

Abstract

In this paper, an analytical method for tuning the parameters of the set-point weighted fractional order PID (SWFOPID) controller is proposed. The studied control scheme is the filtered fractional set-point weighted (FFSW) structure. Also to achieve a desired closed-loop performance, a fractional order pre-filter is employed. The proposed method is applicable to stable plants describable by a simple three-parameter fractional order model. Such a model can be considered as the fractional order counterpart of a first order transfer function without time delay. Finally, the proposed method is implemented on a laboratory scale CE 150 helicopter platform and the results are compared with those of applying a filtered fractional order PI (FFOPI) controller in a similar structure. The practical results show the effectiveness of the proposed method.

Introduction

Reducing undesired oscillations and rapid changes of the control signal is one of the common control objectives. Filtering reference input is an efficient method to reduce undesired oscillations and prevent rapid control signal changes in practical systems [1], [2], [3], [4], [5], [6]. In recent years, fractional order pre-filters have been increasingly used in dynamical systems [7], [8], [9], [10]. The existence of the tunable order α in the structure of the fractional order pre-filters makes these filters more flexible in comparison with the classical pre-filters [7].

This is due to the fact that the fractional calculus has a great potential to improve the traditional methods in different fields of control systems; such as controller design [11], [12], [13], [14], [15], [16], [17], [18], [19], [20] and system identification [21], [22], [23], [24], [25]. Fractional order concepts are employed in simple and advanced control methodologies; such as set-point weighted fractional order PID (SWFOPID) controller [11], [19], phase lead and lag compensator [14], [26], internal model based fractional order controller [27], Smith predictor based fractional order controller [18], [21], optimal fractional order controller [28], [29], robust fractional order control systems [30], [31], [32], fuzzy fractional order controller [33], fractional order sliding mode controller [34], fractional order switching systems [35], and adaptive fractional order controller [17], [36]. It has been shown that the fractional order controllers are more effective in some industrial applications [12], [15], [20]. Also, this type of controllers have been applied in a wide range of practical applications; such as control of magnetic flux in tokamak machine [19], heat flow platform [20], hexapod robot [37], permanent magnetic synchronous motors [38], and power electronic buck converters [39].

Fractional order models have been used in modelling different physical systems; such as modelling of lead-acid battery [40], human arm [41], dynamic backlash [42], thermal system [43], and large three-dimensional RC networks [44]. One significant motivation for applying fractional order models is that these models can well approximate the dynamics of high order classical models [45]. In this paper, it is shown that the fractional order counterpart of a first order transfer function without time delay can be a good candidate for characterizing the behaviour of the CE 150 helicopter which is inherently a high order plant.

The purpose of this paper is to propose an analytical method for tuning fractional order PID (FOPID) controllers based on the set-point weighted control structure. The set-point weighted structure is a simple and widely used control strategy for compensating stable, integrative, and unstable systems [11], [46], [47]. The advantage of the set-point weighted control structure is that the system responses to disturbances and set-point changes can be adjusted separately [47]. Different strategies are available for tuning the parameters of the set-point weighted PI and PID controllers [46], [47], [48]. Also in recent years, different methods have been proposed for designing the parameters of the SWFOPID controllers [11], [19]. In [11], an effective way is proposed to tune the set-point weights when the FOPID controller is designed to optimize the disturbance rejection performance based on the standard fractional set-point weighted control structure. In [19], a simple analytical method is proposed for tuning the parameters of the SWFOPID controller based on the standard fractional set-point weighted control structure.

The main contribution of this paper is to propose a simple analytical method for designing FOPID controllers based on the filtered fractional set-point weighted (FFSW) control structure. In this method, a fractional order pre-filter is employed to achieve a desired closed-loop performance. In order to evaluate the proposed method, the designed SWFOPID controller is implemented on a laboratory scale CE 150 helicopter platform and the results are compared with those of applying a filtered fractional order PI (FFOPI) controller in a similar structure. Also, a double feedback loop method is used to achieve stability and better system performance. Effectiveness and simplicity are key features of the presented method.

This paper is organized as follows. Section 2 presents some preliminaries on fractional calculus, FOPID controllers, and stability of fractional order transfer functions. The analytical design procedure of the FOPID controller based on the FFSW control structure is described in Section 3. In Section 4, the analytical design procedure of the FFOPI controller is described. The sensitivity functions of the designed control systems are compared in Section 5. In Section 6, the CE 150 helicopter platform is introduced and experimental results by applying the designed FOPID controller based on the FFSW control structure in comparison with the FFOPI controller are provided. Finally, the paper is concluded in Section 7.

Section snippets

Preliminaries

Caputo fractional derivative is the most frequently used definition of fractional order derivative [49]. The Caputo definition of the fractional order derivative of order γ, where γ is a positive non-integer number is given byDtγacx(t)=1Γn-γatxnτt-τγ-n+1dτ,n-1<γ<nNxn(t),γ=n.In (1), Γγ is the Gamma function. Also, a and t are the lower and the upper terminals of the integral [49]. The Laplace transform of the Caputo fractional derivative is given byL{Dtγacx(t)}=sγL{x(t)}-j=0n-1sγ-1-jx(j)(0),n-

Fractional set-point weighted PID controller design

The general schematic of a set-point weighted control structure is shown in Fig. 2. In this structure, P(s) is the process transfer function. The process is identified by a simple three-parameter fractional order model. The control scheme studied in this paper is the FFSW structure shown in Fig. 3. In this structure to achieve a desired closed-loop response, an FOPID controller is employed. The set-point filter Gf(s) is used to reduce undesired oscillations. In the next subsection, an

Filtered fractional order PI controller design

In this section, an analytical method is proposed to design FFOPI controllers based on the control structure shown in Fig. 5. The closed-loop response of the proposed method in this section is similar to the results of applying the SWFOPID controller described in Section 3. At the end of the next section, it is shown that disturbance rejection of the SWFOPID controller is always better than the FFOPI controller.

Comparison of the designed controllers

In order to explain the difference of two proposed method in this paper, the sensitivity functions of the designed control systems are compared in this section.

Based on relations (16), (17), (20), (21), by applying the SWFOPID controller in the control structure shown in Fig. 4, the sensitivity function S(s)=1/1+Gc(s)P(s) and the complementary sensitivity function T(s)=Gc(s)P(s)/1+Gc(s)P(s) are respectively obtained asSSWFOPID(s)=P2-P1sαP2sα+1,andTSWFOPID(s)=P1sα+1P2sα+1.

Also according to

Experimental results

In this section, the CE 150 laboratory scale helicopter is selected for experimental application of the designed fractional order controllers. The system is shown in Fig. 6 and different control strategies have been applied to this laboratory helicopter [54], [55], [56], [57], [58]. In [55], [57], [58], high order classical models have been used for modelling the CE 150 helicopter. As previously mentioned, fractional order models can be good candidate for describing the dynamics of systems

Conclusions

In this paper, a simple analytical method for tuning the parameters of FOPID controller based on the FFSW control structure was proposed. The parameters of the FOPID controller, set-point weight, and constant coefficient of the fractional order set-point filter were tuned to achieve the desired closed-loop transfer function. Also, it was shown that disturbance rejection by applying the SWFOPID controller is always better done in comparison with the designed SWPID, FFOPI and FPI controllers in

References (60)

  • H. Malek et al.

    Identification and tuning fractional order proportional integral controllers for time delayed systems with a fractional pole

    Mechatronics

    (2013)
  • M. Tavakoli-Kakhki et al.

    Simple fractional order model structures and their applications in control system design

    Eur J Control

    (2010)
  • S. Victor et al.

    Parameter and differentiation order estimation in fractional models

    Automatica

    (2013)
  • D. Valério et al.

    Identifying a non-commensurable fractional transfer function from a frequency response

    Signal Process

    (2015)
  • M. Tavakoli-Kakhki et al.

    Proportional stabilization and closed-loop identification of an unstable fractional order process

    J Process Control

    (2014)
  • M. Tavakoli-Kakhki et al.

    Fractional order model reduction approach based on retention of the dominant dynamics: application in IMC based tuning of FOPI and FOPID controllers

    ISA Trans

    (2011)
  • A. Hajiloo et al.

    Pareto optimal robust design of fractional-order PID controllers for systems with probabilistic uncertainties

    Mechatronics

    (2012)
  • F. Padula et al.

    Tuning rules for optimal PID and fractional-order PID controllers

    J Process Control

    (2011)
  • D.J. Wang et al.

    H Design with fractional-order PDμ controllers

    Automatica

    (2012)
  • H. Özbay et al.

    PID controller design for fractional-order systems with time delays

    Syst Control Lett

    (2012)
  • S. Das et al.

    A novel fractional order fuzzy PID controller and its optimal time domain tuning based on integral performance indices

    Eng Appl Artif Intell

    (2012)
  • C. Yin et al.

    Fractional-order sliding mode based extremum seeking control of a class of nonlinear systems

    Automatica

    (2014)
  • S.H. HosseinNia et al.

    Stability of fractional order switching systems

    Comput Math Appl

    (2013)
  • Y. Luo et al.

    Fractional order robust control for cogging effect compensation in PMSM position servo systems: stability analysis and experiments

    Control Eng Pract

    (2010)
  • A.J. Calderón et al.

    Fractional order control strategies for power electronic buck converters

    Signal Process

    (2006)
  • J. Sabatier et al.

    A fractional order model for lead-acid battery crankability estimation

    Commun Nonlinear Sci Numer Simul

    (2010)
  • I. Tejado et al.

    Fractional order human arm dynamics with variability analyses

    Mechatronics

    (2013)
  • A. Maachou et al.

    Nonlinear thermal system identification using fractional Volterra series

    Control Eng Pract

    (2014)
  • G. Prashanti et al.

    Set-point weighted PID controllers for unstable systems

    J Franklin Inst

    (2000)
  • H. Boubertakh et al.

    Tuning fuzzy PD and PI controllers using reinforcement learning

    ISA Trans

    (2010)
  • Cited by (49)

    • Fault-tolerant optimal pitch control of wind turbines using dynamic weighted parallel firefly algorithm

      2022, ISA Transactions
      Citation Excerpt :

      Similar to PID controllers, fractional-order PID (FOPID) controllers have been extensively implemented in many applications [18–20]. FOPID controllers not only inherit the advantages of conventional PIDs such as simple structure and strong robustness but also expand the control range by adding more flexibility to the control system [15,18]; however, the existence of two more tunable parameters has made the design problem more complicated. A variety of tuning rules and design methods have been investigated in the literature [15,21], while most of them suffer from the unavailability of the exact dynamic model in Laplace domain representation, especially for complex nonlinear systems [18].

    • A robust fractional-order controller design with gain and phase margin specifications based on delayed Bode's ideal transfer function

      2022, Journal of the Franklin Institute
      Citation Excerpt :

      The usefulness of fractional-order models in representing the higher integer-order transfer functions arouses from the fact that they possess infinite number of zeros and poles even though they are represented by a few fractional parameters. Fractional-order models represent system behavior much better than integer-order counterparts in various real-time system applications such as liquid level system [2], heat flow platform [11], twin rotor helicopter [12], servo-system [13,14]. In most of these studies, the controllers are designed utilizing the simplest fractional-order model formed by a fractional order insertion to the first integer-order model.

    • Fractional order MIMO controllers for robust performance of airplane longitudinal motion

      2019, Aerospace Science and Technology
      Citation Excerpt :

      Because of the success in the synthesis of real fractional order differentiators [7], and the robustness and performance of fractional order controllers, these controllers have been designed and applied to control a large variety of dynamical processes, including both integer order and fractional order systems [8,9]. Moreover, because of the robust performance and fast response, many aerospace engineers have tended to use fractional order controllers to control aerospace systems such as spacecraft [10], quadrotor [11], servo actuation system [12], space tether [13], helicopter [14], hose–drogue system [15], and space manipulator [16]. Nonetheless, to our best knowledge, no previous attempt has been made to control of airplanes, which play an important role in all types of transportation.

    • Analytical fractional PID controller design based on Bode's ideal transfer function plus time delay

      2019, ISA Transactions
      Citation Excerpt :

      Various studies have proven that reduced fractional-order models represent system behavior much better than integer order counterparts [1,13]. In literature, the reduced fractional order models have been used in various controller design methodologies to induce more flexibility [1,4,9–13]. A fractional order PID design method for higher order processes is addressed in [13].

    View all citing articles on Scopus
    View full text