Inter national J our nal of Electrical and Computer Engineering (IJECE) V ol. 8, No. 6, December 2018, pp. 4282 4289 ISSN: 2088-8708, DOI: 10.11591/ijece.v8i6.pp4282-4291 4282       I ns t it u t e  o f  A d v a nce d  Eng ine e r i ng  a nd  S cie nce   w     w     w       i                       l       c       m     A No v el Nonlinear Contr ol of Boost Con v erter using CCM Phase Plane Ek o Setiawan 1 and Ichijo Hodaka 2 1 Interdisciplinary Graduate School of Agriculture and Engineering, Uni v ersity of Miyazaki, Japan 2 Department of En vironmental Robotics, Uni v ersity of Miyazaki, Japan Article Inf o Article history: Recei v ed Jan 20, 2018 Re vised Jul 18, 2018 Accepted Aug 7, 2018 K eyw ord: Nonlinear control Boost con v erter Continuous Conduction Mode Phase plane ABSTRA CT Boost con v erter is one of fundamental DC-DC con v erters and used to deli v er electric po wer with boosted v oltage in man y electrical systems. Se v eral control strate gies ha v e been applied to control a boost con v erter deli v ering a constant output v oltage. Gener - ally , boost con v erter w orks in tw o modes; one is called a Continuous Conduction Mode (CCM). Man y researches use CCM model in the controller design, b ut the y ne v er en- sure that the controller al w ays w orks in CCM. This paper proposes no v el nonlinear controller of boost con v e rter designed using the modification of flo w in phase plane. The proposed controller guarantees that the boost con v erter w orks only in CCM re gion. The simulation result confirms that our proposed controller brings the state v ariables from an y initial point to a desired operating point successfully . Copyright c 2018 Institute of Advanced Engineering and Science . All rights r eserved. Corresponding A uthor: Ichijo Hodaka Department of En vironmental of Robotics, Uni v ersity of Miyazaki, Japan 1-1, Gakuen Kibanadai-nishi, Miyazaki, 889-2192, Japan Phone: +81 985 587 352 Email: hijhodaka[at]cc.miyazaki-u.ac.jp 1. INTR ODUCTION Ov er the last fe w decades, DC-DC con v erters ha v e been the subject of great interest due to its e xtensi v e increment of utilization in dif ferent applications. Right no w , the y are popularized in standard and redone items that po wer an e xtensi v e v ariety of applications, for e xample, photo v oltaic (PV) po wer systems [1, 2], wind turbines (WT) [3, 4 ] , brushless DC (BLDC) motor [5, 6] etc. Among these con v erters, the boost con v erter is a fundamental controller which is used in man y systems due to its simplicity . In order to achie v e the operating point, boost con v erter usually w orks with the controller techniques . There are tw o controllers for the DC-DC con v erter as pulse-width modulation (PWM) and phase-shift modu- lation (PSM). The PWM has been widely utilized to control of DC-DC con v erter in se v eral applications. In the case of less number of components usage and high-reliability demand, the PWM control sho ws the better performance than PSM [7]. Proportional-inte gral-deri v ati v e (PID) and sliding-mode control (SMC) are used widely in a DC-DC con v erter . The PID control of fers the good stability system, b ut it only operates on limited operating point. The SMC pro vides lar ger operating point than PID. The SMC w orks well at most operating point. Ho we v er , the fundamental barrier for SMC e x ecution is a marv el called ’chattering’. Based on conduction mode, t he DC-DC con v erter is analyzed in tw o modes as continuous-conducti on mode (CCM) and discontinuous-conduction m o de (DCM). The CCM is the most often used in DC-DC con v erter analysis to design a controller . Although it is often chosen, none of the pre vious literature e xamines that their proposed controller w orks only on CCM re gion. Since t he design is based on CCM, the controller and system should w ork only in that re gion or the anal ysis and design may mislead the controller . Moreo v er , the controller of boost con v erter should be able to handle an y operating points with correct design. The rapid de v elopment of the v ery lar ge scale inte grated circuit technology brings digitally controlled DC-DC con v erter as hot topic [8]. Digital processors also ha v e the adv antage of being less susceptible to aging and en vironmental or parameter v ariations. In addition, the processor can monitor the system, perform self- J ournal Homepage: http://iaescor e .com/journals/inde x.php/IJECE       I ns t it u t e  o f  A d v a nce d  Eng ine e r i ng  a nd  S cie nce   w     w     w       i                       l       c       m     Evaluation Warning : The document was created with Spire.PDF for Python.
IJECE I SSN: 2088-8708 4283 Figure 1. Boost con v erter Figure 2. Discrete-time of state diagnostics and tests, and communicate status to a display or a host computer [9]. Implementation of digital controller requires system analysis in discrete-time domain. This paper proposes nonlinear controller which be able to handle an y initial point and k eep the system w orks in CCM. This paper focuses on one type DC-DC con v erter which is boost con v erter b ut the idea has possibility to be applied on other con v erter . The proposed controller design is based on flo w modification of phase plane. The analysis is conducted in discrete time-domain which is required by a digital controller due to the popularity of digital controller . Section 2 e xplains analysis of boost con v erter in di screte time-domain. Digital implementation of controller requires indirectly the discrete time-domain analysis. Section 3 sho ws the control system specificat ion in CCM of boost con v erter . Section 4 proposes the nonlinear control which its design based on phase plane e xamination of an y ini tial condition. Section 6 simulates the system on se v eral initial points. Finally , Section 7 tells the important point of this paper . 2. D YN AMIC MODEL OF BOOST CONVER TER The boost con v erter consist of inductor L , capacitor C , MOSFET M and diode D as sho wn in F igure 1. The circuit equation of boost con v erter is deri v ed as follo ws. V = L _ i ( t ) + v M ( t ) ; v M ( t ) = v D ( t ) + v ( t ) i ( t ) = i M ( t ) + i D ( t ) ; i D ( t ) = C _ v ( t ) + v ( t ) =R (1) The diode v oltage and MOSFET v oltage are denoted as v D and v M respecti v ely . A resisti v e load R is connected to the boost con v erter . A v oltage source V supplies a constant v oltage for boost con v erter . The con v erter is controlled by duty-ratio d of PWM. In CCM, there are tw o modes which w ork alternately . In the first mode or mode-1, the MOSFET M is conducted and the diode D is di sconnected. W e assume that the MOSFET v oltage is constant V M during mode-1 and there is no current on diode ( i D ( t ) = 0 ). The second mode or mode-2 occurs when the MOSFET is of f and diode is conducted. W e assume that the diode v oltage is constant V D and there is no current on MOSFET ( i M ( t ) = 0 ) during mode-2. The equation of boost con v erter becomes as follo ws. mode-1 L _ i ( t ) = V V M C _ v ( t ) = v ( t ) =R (2) mode-2 L _ i ( t ) = V V D v ( t ) C _ v ( t ) = i ( t ) v ( t ) =R : (3) In this section, we introduce a non-dimensional v ariable x 1 and x 2 which is described as follo ws. ( x 1 = v ( t ) V + V D V x 2 = i ( t ) V q L C (4) The deri v ati v e of non-dimensional state on each mode can be calculated by substituting _ i ( t ) and _ v ( t ) of equation (2) and (3) as follo ws. mode-1 ( _ x 1 = _ v ( t ) V = 1 V ( V x 1 + V V D ) R C = 1 R C x 1 1 R C 1 V D V _ x 2 = _ i ( t ) V q L C = 1 V q L C V V M L = 1 p LC 1 V M V (5) mode-2 8 < : _ x 1 = _ v ( t ) V = 1 V i ( t ) C v ( t ) R C = 1 p LC x 2 1 R C x 1 1 R C 1 V D V _ x 2 = _ i ( t ) V q L C = 1 V q L C V V D v ( t ) L = 1 p LC x 1 (6) A No vel Nonlinear Contr ol of Boost Con verter using CCM Phase Plane (Ek o Setiawan) Evaluation Warning : The document was created with Spire.PDF for Python.
4284 ISSN: 2088-8708 The state space equation of non-dimensional v ariable can be written as follo ws. _ x = 1 R C 0 0 0 | {z } A 1 x + " 1 R C (1 V D V ) 1 p LC 1 V M V # | {z } b 1 (7) _ x = " 1 R C 1 p LC 1 p LC 0 # | {z } A 2 x + 1 R C (1 V D V ) 0 | {z } b 2 (8) As sho wn in Figure 2, mode-1 w orks from the be ginning t 0 until t = t 0 + T d . Based on general solution of state space equation, the last state of mode-1 can be written as follo ws. x ( t 0 + T d ) = e A 1 ( t 0 + T d t 0 ) x ( t 0 ) + Z t 0 + T d t 0 e A 1 ( t 0 + T d p ) b 1 dp where 8 < : q = p t 0 T p = q T + t 0 dp dq = T = e A 1 T d x ( t 0 ) + Z d 0 e A 1 T ( d q ) b 1 T dq (9) In CCM, the boost con v erter has tw o state-space equations (7) and (8). The last state of mode-1 is equal to the initial state of mode-2. On the ne xt mode, the last state of mode-2 will be as the initial state of the ne xt period mode-1. These phenomenons occurs repeatedly . Based on these f acts, the solution of boost con v erter per period ( T ) can be obtained by substituting the last state of mode-1 ( x ( t 0 + T d ) ) into the general solution of mode-2 as follo ws. x ( t 0 + T ) = e A 2 T (1 d ) x ( t 0 + T d ) + Z 1 d e A 2 ( t 0 + T ( q T + t 0 )) b 2 T dq = e A 2 T (1 d ) e A 1 T d x ( t 0 ) + e A 2 T (1 d ) Z d 0 e A 1 T ( d q ) b 1 T dq + Z 1 d e A 2 T (1 q ) b 2 T dq (10) Let us assume that the sensor measures e v ery end of switching period ( T ). This measurement is sho wn as dot-point in Figure 2. W e will introduce ne w v ariable to distinguish with the continuous-time v ariable x . Then, the discrete representation of solution (10) is defined as follo ws: ( k + 1) = x ( t 0 + ( k + 1) T ) = e ~ A 2 (1 d ) e ~ A 1 d x ( t 0 + k T ) | {z } ( k ) + e ~ A 2 (1 d ) Z d 0 e ~ A 1 ( d q ) ~ b 1 dq + Z 1 d e ~ A 2 (1 q ) ~ b 2 dq (11) where ~ A 1 = " 1 0 0 0 , ~ A 2 = " 1 " 2 " 2 0 , ~ b 1 = " 1 " 2 , ~ b 2 = " 1 0 , " 1 = T R C , " 2 = T p LC , = (1 V M V ) , and = (1 V D V ) (12) Assuming the period T is small then the element of ~ A 1 , ~ A 2 , ~ b 1 , and ~ b 2 become small too. The e xpo- nential part in (11) can be calculated using the definition of e xponential as follo ws. e M = I + M + 1 2! M 2 + ::: e ~ A 1 d = I + ~ A 1 d + ~ A 1 2 d 2 2 + ::: | {z } ne glected ' I + ~ A 1 d e ~ A 2 (1 d ) e ~ A 1 d ' I + ~ A 1 d + ~ A 2 (1 d ) e ~ A 2 (1 d ) Z d 0 e ~ A 1 ( d q ) ~ b 1 dq ' ( I + ~ A 2 (1 d )) Z d 0 ( I + ~ A 1 ( d q )) ~ b 1 dq = ~ b 1 d Z 1 d e ~ A 2 (1 q ) ~ b 2 dq ' Z 1 d ( I + ~ A 2 (1 q )) ~ b 2 dq = ~ b 2 (1 d ) (13) IJECE V ol. 8, No. 6, December 2018: 4282 4289 Evaluation Warning : The document was created with Spire.PDF for Python.
IJECE I SSN: 2088-8708 4285 Thus, the discrete-time solution of state (11) is simplified as follo ws ( k + 1) = ( I + ~ A 1 d + ~ A 2 (1 d )) | {z } ^ A ( k ) + ~ b 1 d + ~ b 2 (1 d ) | {z } ^ b (14) The equation (14) is used to simulate the boost con v erter on the ne xt sections. 3. CONTR OL PR OBLEM A boost con v erter in continuous-conduction mode (CCM) has the limitation such as: 1. the output v oltage v ( t ) must be greater than the input v oltage V subtracted by the diode v oltage V D . This is the principle of boost. In the non-dimensional state, 1 ne v er be ne g ati v e, 2. the inductor current i ( t ) must be greater than zero as the definition of CCM. It means that the state 2 must not ne g ati v e, and 3. the tw o pre vious conditions are satisfied for an y initial states. Breaking the limitation means the model is not proper an ymore in the controller design. K eeping the wrong model may mislead the analysis of controller design. The three limitations will be used as control specification in this paper . T able 1. The parameter of Boost Con v erter P arameter Symbol V alue Input v oltage V 10 V Inductor L 300 H Capacitor C 100 F MOSFET v oltage V M 162 mV Diode v oltage V D 0.5 V Load R 10 Switching period T 20 s Reference v oltage V r ef 16 V Let us s imulate the beha vior of boost con v erter using parameters of boost from [12] as sho wn in T able 1. The param eters are also used in the ne xt sections. W e e xamines the beha vior of boost con v erter without the controller called as open-loop response. The boost con v erter is gi v en a constant equilibrium duty ratio notated as D eq . The mathematics softw are (e.g. W olfram Mathematica) finds the v alue of equilibrium duty-ratio ( D eq ) by solving the duty-ratio when equation (14) is equal to [ 1 ; 2 ] T as follo ws. D eq = r ef + r ef (15) where r ef = ( V r ef V + V D ) =V . In order to e xamine the beha vior of boost for an y initial condition, let us utilize phase pl ane. The phase plane of open-loop response is sho wn in Figure 3. The dashed line i n Figure 3 sho ws the set of equilibrium condition. Let us focus on the phase trajectory of tw o initial conditions which are notated as A and B. It sho ws that the system can achie v e equilibrium point well for both initial conditions. F or the initial condition B, the system enters the ne g ati v e re gion of 1 and 2 which breaks the limitation of CCM boost con v erter . In order to a v oid this situation, a proper controller is required. 4. PID CONTR OLLER Among the se v eral controllers of boost con v erter , PID controller is the mature controller which well- e xplained in se v eral papers such as [7, 8, 11, 13, 14]. This section discusses the implementation of PID controller and its characteristic on boost con v erter . The PID controller i n discrete-time domain is e xpressed as follo ws [11]. u ( k ) = K P 2 4 e ( k ) + T T I k X j =0 e ( j ) + T D T f e ( k ) e ( k 1) g 3 5 (16) A No vel Nonlinear Contr ol of Boost Con verter using CCM Phase Plane (Ek o Setiawan) Evaluation Warning : The document was created with Spire.PDF for Python.
4286 ISSN: 2088-8708 Figure 3. Phase plane of open-loop response The recursi v e e xpression of PID control in discrete-time is formed by dif ference between simultaneous input ( u ( k ) = u ( k ) u ( k 1) ). The pre vious PID control can be e xpressed as follo ws [11, 12]: u ( k ) = u ( k 1) + u ( k ) = d ( k 1) + ( K P + K I + K D ) e ( k ) ( K P + 2 K D ) e ( k 1) + K D e ( k 2) (17) where K I = K P T T I , K D = K P T D T , e ( k ) = V r ef v ( k ) , e ( k 1) = V r ef v ( k 1) , and e ( k 2) = V r ef v ( k 2) . PID controller or called com p e nsator in some references is usually designed by small-signal model of DC-DC con v erter [10]. The small-signal model is deri v ed by adding small pert urbation on inductor current i ( t ) , capacitor v oltage v ( t ) and duty-ratio d . Since the PID controller is designed by linearizat ion around the equilibrium point, implementation of PID controller needs equilibrium duty-ratio D eq as described in follo wing equation [13]. d ( k ) = D eq + u ( k ) (18) Let us e xamine the beha vior of PID controller in the phase plane. The parameter of PID needs to be tuned before used. Based on [14], the Zie gler -Nichols (ZN) has the best performance comparing with the others. The simulation sho ws that the boost system achie v es ultimate g ain ( K U ) and ultimate period ( T U ) on 0.06 and 1,8 ms respecti v ely . According to the ZN table on [14], the P-g ain ( K P ), D-g ain ( K D ), and I-g ain ( K I ) are 0.036, 8 10 4 and 0.405 respecti v ely . Let us dra w the phase trajectory of PID controller on se v eral initial condition v (0) s and i (0) s. The phase trajectory of PID controller is observ ed by applying (14), (17) , (18) and (15) on the se v eral initial points. Figure 4 sho ws the phase trajectory of PID controller for se v eral initial conditions. Based on Figure 4, the flo w of state tends to go to ne g ati v e area of 2 at first. Most of tested initial state enters the ne g ati v e area of 1 and 2 which breaks the limitation of Boost con v erter in CCM . This f act sho ws that the PID does not guarantee the boost con v erter w orks al w ays in CCM. It means that the controller design using CCM is not suitable with the implementation. 5. PR OPOSED CONTR OLLER This paper proposes the nonlinear feedback control which is des igned based on manipulation of flo w i n phase plane. The flo w is consisted from tw o v ectors which are ( 1 ( k ) 1 ( k 1) ) and ( 2 ( k ) 2 ( k 1) ) as sho wn in Figure 5. The flo w is forced to has a specific direction. In that condition, the follo wing equality w orks. ( 1 ( k ) 1 ( k 1)) cos + ( 2 ( k ) 2 ( k 1)) sin = 0 (19) The proposed controller is designed by solvi n g the duty ratio d in the equation (19). Then, a proportional controller is added to push the controller to the reference r ef = ( v r ef V + V D ) =V . Finally , the o v erall proposed controller is described as follo ws. d pr oposed ( k ) = k ( r ef 1 ( k )) + " 2 1 ( k ) sin ( " 1 + " 1 1 ( k ) " 2 2 ( k )) cos " 2 [ 2 ( k ) cos + ( + 1 ( k )) sin ] (20) The proposed controller consists of tw o parts. The firs t part is proportional term and the second part is modification of flo w . The parameter k and needs to be tuned, which represent speed of achie v ement reference IJECE V ol. 8, No. 6, December 2018: 4282 4289 Evaluation Warning : The document was created with Spire.PDF for Python.
IJECE I SSN: 2088-8708 4287 (a) v (0) = 11 V ; i (0) = 1 A (b) v (0) = 11 V ; i (0) = 4 A (c) v (0) = 20 V ; i (0) = 1 A (d) v (0) = 20 V ; i (0) = 4 A Figure 4. Phase trajectory of PID controller Figure 5. Modification of flo w Figure 6. Phase plane of proposed controller ( V r ef = 16 V) and flo w direction respecti v ely . The best v alue of k and are 0.06 and (-0.35 ) respecti v ely . The phase plane of proposed controller is sho wn in Figure 6. The phase plane sho ws that the controller can handle an y initial point on CCM boost con v erter . An y initial points are pushed into the equilibrium line (dashed line in Figure 3) without passing the ne g ati v e area of 1 and 2 . 6. SIMULA TION This section simulates the proposed controller response on time domain. The proposed controller (20) is compared with PID controller (18) which described in Section 4. The simulation is conducted in se v eral initial points. The comparison between proposed and PID controller are sho wn in Figure 7. The PID controller response is sho wn in dashed-line while the proposed controller is e xpressed in solid blue line. The proposed controller sho w smother response than PID controller . In the PID controller , the output v oltage goes to less than input v oltage 10 V . This condition is not proper for boost con v erter characteristic which must greater than the input v oltage. Moreo v er , the phase plane of proposed controller on v arious reference point is sho wn in Figure 8. Figure 8 gi v es clear information that the proposed controller has capability to handle an y references point without entering the ne g ati v e area of states. A No vel Nonlinear Contr ol of Boost Con verter using CCM Phase Plane (Ek o Setiawan) Evaluation Warning : The document was created with Spire.PDF for Python.
4288 ISSN: 2088-8708 (b) v (0) = 11 V ; i (0) = 1 A (c) v (0) = 11 V ; i (0) = 4 A (d) v (0) = 20 V ; i (0) = 1 A (e) v (0) = 20 V ; i (0) = 4 A Figure 7. Comparison between proposed and PID on se v eral initial points (a) v r ef = 9 : 5 V (b) v r ef = 13 : 0 V (c) v r ef = 16 : 5 V (d) v r ef = 20 : 0 V (e) v r ef = 23 : 5 V (f) v r ef = 27 : 0 V Figure 8. Proposed phase plane of the v arious reference point 7. CONCLUSION This paper has pointed out that the PID control does not guarantee that boost con v erter al w ays w orks within CCM. This paper has formulated a discrete approximation of CCM boost con v erter , and proposed a IJECE V ol. 8, No. 6, December 2018: 4282 4289 Evaluation Warning : The document was created with Spire.PDF for Python.
IJECE I SSN: 2088-8708 4289 no v el nonlinear controller that guarantees that boost con v erter al w ays w orks properly within CCM for an y initial condition. Moreo v er , the proposed controller has adv antage more than PID that it can handles an y reference point without tuning the parameters ag ain. REFERENCES [1] S.D. Stallon, K.V . K umar , S.S. K umar , ”High Ef ficient Module of Boost Con v erter in PV Module, International Journal of Electrical and Computer Engineering , v ol. 2, pp. 758-781, 2012. [2] S. Bhat, H.N. Nag araja, ”Ef fect of P arasitic Elements on the Performance of Buck-Boost Con v erter for PV Systems, International Journal of Electrical and Computer Engineering , v ol. 4, pp. 831-836, 2014. [3] A. Boulahia, K. Nabti, H. Benalla, ”Direct Po wer Control for A C/DC/A C Con v erters in Doubly Fed Induction Generators Based W ind T urbine, International Journal of Electrical and Computer Engineering , v ol. 2, pp. 425-432, 2012. [4] T .Z. Khaing, L.Z. K yin, ”Control Analysis of Stand-Alone W ind Po wer Supply System with Three Phase PWM V oltage Source In v erter and Boost Con v erter , International Journal of Electrical and Computer Engineering , v ol. 5, pp. 798-809, 2015. [5] V . Ramesh, Y .K. Latha, ”An Interlea v ed Boost Con v erter Based PFC Control Strate gy for BLDC mot or , International Journal of Electrical and Computer Engineering , v ol. 5, pp. 957-966, 2015. [6] T . Raghu, S.C. Sekhar , J.S. Rao, ”SEPIC Con v erter based-Dri v e for Unipolar BLDC Motor , International Journal of Electrical and Computer Engineering , v ol. 2, pp. 159-165, 2012. [7] M.Z. Hossain, N.A. Rahim, J.a. Selv araj, ”Recent progress and de v elopment on po wer DC-DC con v erter topology , control, design and applications: A re vie w , Rene w able and Sustainable Ener gy Re vie ws , v ol. 81, pp. 205-230, 2017. [8] Z. Shen, N. Y en, H. Min, ”A Multimode Digitally Controlled Boost Con v erter with PID Autotuning and Constant Frequenc y/ Constant Of f-time Hybrid PWM Control, IEEE T ransaction on Po wer Electronics , v ol. 26, pp. 2588-2598, 2011. [9] L.Guo, M.Aqil, D.S. Zinger , J. W ang, ”Design of a digital control system for DC-DC con v erter to po wer electromagnets, IEEE Industry Applications Society Annual Meeting , pp. 1-5 2013. [10] R. W . Erickson, D. Maksimo vic, ”Fundamental of Po wer Electronics, Kluwer Academic Publishers, Netherland, 2001. [11] S. Chander , P . Ag arw al, I. Gupta, ”FPGA-based PID Controller for DC-DC Con v erter , IEEE Joint International Con- ference on Po wer Electronics, Dri v es and Ener gy Systems & Po wer India , 2010. [12] M.M. Peretz, S. Ben-Y aak o v , ”T ime-Domain Design of Digital Compensators for PWM DC-DC Con v erters, IEEE T ransaction on Po wer Electronics , v ol. 27, pp. 284-293, 2012. [13] Y .I. Son, I.H. Kim, ”Complementary PID Controller to P assi vity-Based Nonlinear Control of Boost Con v erters W ith Inductor Resistance, IEEE T ransactions on Control Systems T echnology , v ol. 20, pp. 826-834, 2012. [14] O. Ibrahim, N.Z. Y ahaya, N. Saad, ”Comparati v e studies of PID controller tuning methods on a DC-DC boost con- v erter , IEEE International Conference on Intelligent and Adv anced Systems (ICIAS) , August 2016. A No vel Nonlinear Contr ol of Boost Con verter using CCM Phase Plane (Ek o Setiawan) Evaluation Warning : The document was created with Spire.PDF for Python.