Transient Stability Enhancement of Variable Speed Permanent Magnet Wind Generator using Adaptive PI-Fuzzy Controller

Transcription

1 This is the author's version of the paper published in this conference. Changes may appear in the published version. DOI: 1.119/PTC Transient Stability Enhancement of Variable Speed Permanent Magnet Wind Generator using Adaptive PI-Fuzzy Controller Marwan Rosyadi, Student Member, IEEE, S.M. Muyeen, Member, IEEE, Rion Takahashi, Member, IEEE, and Junji Tamura, Senior Member IEEE Abstract This paper proposes an application of adaptive PI- Fuzzy controller for back to back converter of variable speed permanent magnet wind generator in order to enhance transient stability of wind farms including induction generators. The back to back converter control scheme is developed based on the field orientation control, in which both active and reactive power delivered to a power grid system are controlled effectively. The adaptive PI-Fuzzy controller is proposed to enhance performances of the variable speed permanent magnet wind generator in various operating conditions. To evaluate the controller system capabilities, simulation analyses are performed on a model system composed of a wind farm connected to an infinite bus. The simulations have been performed using PSCAD/EMTDC. Simulation results show that the proposed control scheme is more effective to enhance the transient stability of wind farm during severe network disturbance compared with that with conventional PI controller. Index Terms Variable speed wind turbine, permanent magnet synchronous generator, back to back converter, PI controller, fuzzy logic controller. I. INTRODUCTION ecently the penetration of wind power generators into Relectrical grids has significantly increased. Induction generators (IGs) are used widely as a fixed speed wind generator due to their superior characteristics such as brushless and rugged construction, low cost, and operational simplicity [1]. It is well known that traditional fixed speed wind turbine (FSWT) generators directly connected to the grid do not have high control capabilities and cannot guarantee low voltage ride-trough capability for voltage dips []. The squirrel cage induction generators require large reactive power to recover the air gap flux when a short circuit fault occurs in the grid system [3]. If sufficient reactive power is not supplied, the electromagnetic torque of the induction generator decreases significantly and the difference between mechanical and electromagnetic torques becomes large. Then the induction generator and also wind turbine speeds increase rapidly. As a result, the induction generator becomes unstable and it requires to be disconnected from the grid system. Voltage source or current source inverter based Flexible AC Transmission System (FACT) devices such as Static Var All authors are with the department of Electrical and Electronic Engineering, Kitami Institute of Technology, 15 Koen-cho, Hokkaido, Kitami, 9-857, Japan ( marwanrosyadi@yahoo.co.id). Compensator (SVC), Static Synchronous Compensator (STATCOM), Dynamic Voltage Restorer (DVR), Solid State Transfer Switch (SSTS), and Unified Power Flow Controller (UPFC) have been used for flexible power flow control, secure loading and damping of power system oscillations [-]. However, installation of such FACTS devices at a fixed speed wind generator based wind farm increases the system overall cost. The variable speed wind turbine (VSWT) has recently become the dominant type among installed wind turbines. VSWT is designed to achieve maximum aerodynamic efficiency over the wide range of wind speed, increase energy capture, improve power quality and reduce mechanical stress on the wind turbine [7]. Variable Speed Wind Turbine with Permanent Magnet Synchronous Generator (VSWT-PMSG) and back to back converter is considered to be a promising wind turbine concept [8]. The advantages of VSWT-PMSG configuration are: 1) No Gearbox and no brushes, and thus higher reliability; ) The full power converter totally decouples the generator from the grid, and hence grid disturbances have no direct effect on the generator; 3) No additional power supply for magnetic field excitation is needed; ) The converter permits very flexible control of active and reactive power in cases of normal and disturbed grid conditions; 5) The amplitude and frequency of the generator voltage can be fully controlled by the converter [9,1]. VSWT-PMSG system equipped with full rating power electronic converters has strong fault ride through capability during a network disturbance. Moreover, the VSWT-PMSG system with fully scale converter not only can recover a network voltage drop preventing instability of Fixed Speed Wind Turbines with IGs (FSWT-IG) when fault occurs, but also generates electric power in steady state. Therefore, installation of VSWT-PMSG with IGs in a wind farm can be considered to be much effective. VSWT-PMSG has been proposed also to enhance transient stability of wind farm including FSWT-IG, because it can control reactive power as well as provide maximum power to the grid [11]. PI controller is very common in the control of back to back converter of PMSG. However, one disadvantage of this conventional control is the fact that the controller with using fixed gains may not provide the required control performance when there are variations in the system parameters or operating conditions [1]. In order to improve control performances of PMSG, online adjusted adaptive gain method

2 based on Fuzzy Logic Control (FLC) is proposed in this paper. Hence transient stability of overall wind farms can be enhanced more effectively. II. MODEL SYSTEM The model system used in this study is shown in Fig. 1. Here, Wind Farm 1 with PMSG (5. MVA) is connected to an infinite bus through a frequency converter, a 1./ kv step up transformer and a double circuit transmission line. In the figure, the double circuit transmission line parameters are numerically shown in the form of R+jX, where R and X represent the resistance and reactance, respectively. Another wind farm with IG rated at 5. MVA (Wind Farm ) is connected to the network via a.9/ kv transformer and a short transmission line. A capacitor bank is used for reactive power compensation at steady state. The value of capacitor C is chosen so that the power factor of the wind power station becomes unity during steady state condition. Parameters of PMSG and IG are shown in Table I. β = λ opt β =15 β = β =1 β = 5 Fig.. Cp - λ characteristic for different pitch angle Fig. 3. Turbine power characteristic (β= ) IG Fig. 1. Model system TABLE I GENERATOR PARAMETERS MVA 5 MVA 5 R1.1 (pu) Rs.1 (pu) X1.1 (pu) Xs. (pu) Xm 3.5 (pu) Xd.9 (pu) R1.35 (pu) PMSG Xq.7 (pu) R.1 (pu) Field Flux 1. (pu) X1.3 (pu) H 3. s X.89 (pu) H 1.5 s III. VSWT-PMSG CONTROL SYSTEM A. Wind Turbine Modeling The mathematical relation for mechanical power extraction from the wind can be expressed as follows [13]:.5, (1) where P w is the extracted power from the wind, ρ is the air density (Kg/m3), R is the blade radius (m), Vw is the wind speed, Cp is the power coefficient which is a function of tip speed ratio λ and blade pitch angle β (deg). C p for given values of λ and β is calculated for both fixed and variable speed wind turbines by the following equations, based on the turbine characteristics in [13]:, () with (3) 1 The coefficients c 1 to c are: c 1 =.517, c = 11, c 3 =., c = 5, c 5 = 1 and c =.8. The C p -λ characteristics for different values of the pitch angle β are illustrated in Fig.. The maximum value of C p (C p_opt =.8) is achieved for β = and λ = 8.1. This particular value of λ is defined as the optimal value (λ opt ). Fig. 3 depicts the turbine output power as a function of the rotor speed for different wind speeds with the blade pitch angle β =. This figure is obtained with the default parameters (base wind speed = 1 m/sec, maximum power at base wind speed = 1 pu, and base rotational speed = 1 pu). In variable speed wind turbines, the rational speed (ω r ) of wind turbine is controlled to follow the Maximum Power Point Trajectory (MPPT). Since the precise measurement of wind speed is difficult, it is better to calculate the maximum power P max without measuring the wind speed as follows:.5 _ () where λ opt and C p_opt are optimum values of tip speed ratio and power coefficient, respectively. From eq.(), it is clear that the maximum power generated is proportional to the cube of the rotational speed. To control the generator side converter, the maximum power P max is calculated based on the MPPT, which becomes the reference power for the converter.

3 3 Fig.. Proposed control scheme for VSWT-PMSG B. PMSG Proposed Control Detail of the control scheme for VSWT-PMSG system proposed in this paper is shown in Fig.. The full-rating power converter is made up of two level back-to-back IGBT bridges (the generator side converter and the grid side converter) linked by a DC bus. The three-phase AC output of the generator is rectified, and the DC output from the rectifier is fed to an IGBT-based grid side converter. The grid side converter output is then supplied to a step-up transformer, which is connected to the grid. The control signal used for switching operation of the converters is three-phase sinusoidal voltage. The method for the transformation from d and q axis signal to the three phase sinusoidal signal is also illustrated in Fig., in which the d and q axis reference voltages are supplied from generator side controller (V ds and V qs ) and grid side controller (V dg and V qg ). The angle θ r for the transformation between abc and dq variables of stator current is calculated from the rotor speed of PMSG. The angle θ t for the transformation of grid current from abc to dq variables is detected from the three phase voltage at the high voltage side of the transformer. The grid side voltage phasor is synchronized with the controller reference frame by using Phase Lock Loop (PLL). In both converters, the triangle signal is used as the carrier wave of Pulse Wave Modulation (PWM) operation. The Carrier frequency is chosen 1kHz for both converters. The DC-link capacitor value is 5 μf. The rated DC-link voltage is kv. Generator side converter is directly connected to PMSG, which includes two PI controllers (PI 1 and PI ) and an adaptive PI-Fuzzy Controller (Adaptive PI-Fuzzy 1). The control strategy is based on controlling the active power of the generator by the q-axis component of the stator current and controlling the reactive power by the d-axis current. The q-axis current (I qs ) can control the active power (Pm). The active power reference (P m_ref ) is determined from Maximum Power Point Tracking (MPPT). The d-axis current (I ds ) can control the reactive power (Q m ). The reactive power reference (Q m_ref ) is set to zero for unity power factor operation. The grid side converter works to maintain the DC-link capacitor voltage at the set value, which assures the active power exchange between PMSG and the grid. The converter also control the reactive power output to the grid in order to control the grid side voltage. The controller strategy includes three PI controllers (PI 3, PI, and PI 5) and an adaptive PI- Fuzzy Controller (Adaptive PI-Fuzzy ). DC voltage of the DC-link capacitor (V dc ) is controlled independently via grid d- axis current component (I dq ). On the other hand, the grid q- axis current component (I qg ) can control reactive power of grid side converter (Q g ). The reactive power reference (Q g_ref ) is set so that the terminal voltage at the high voltage side of the transformer remains constant. The DC-chopper is also embedded in the DC-link circuit. When a fault occurs in the grid, a voltage dip appears at the terminal of wind generator and then the active power Pg delivered to the grid is also reduced. As the generator side converter is decoupled with the grid, however, the generator continues to generate the active power Pm and thus the DClink voltage increases due to the power imbalance between the generator side converter and the grid side converter. In order to protect the DC-link circuit from the voltage increase, the DC chopper circuit is proposed in this paper to dissipate the excess energy. For d-axis and q-axis current components regulation, in this paper two adaptive PI-Fuzzy controllers are applied to the generator side converter controller and the grid side converter controller. The control strategy for the PI-Fuzzy controller will be explained in Section IV. IV. ADAPTIVE PI-FUZZY CONTROLLER In industrial applications, PI controllers are most widely used because of their simple structure and good performances in a wide range of operating conditions. In fixed gain controllers, these parameters are selected by methods such as the Zeigler and Nichols, pole placement, etc. These PI controllers are simple but cannot always effectively control systems with changing parameters or strong nonlinearities, and they may need frequent online retuning of their parame-

4 ters. To solve this problem, an adaptive PI-Fuzzy controller is proposed in this paper. The controller is composed of PI controller and fuzzy controller. According to the error and change in error of the control system, the fuzzy controller can online adjust the two parameters of the PI controller in order to be adapted to any variations in the operating conditions. Block diagram of adaptive PI-Fuzzy controller is shown in Fig. 5. The controller is used as current regulator of generator side converter as well as grid side converter. Here four FLCs adjust the PI parameters according to operating conditions, e.g., the error (e) and change in error (de) of the input signals, which characterizes its first time derivative during process control. To determine control signal for proportional gain (Kp) and integration gain (Ki), inference engine with rule base having if-then rules in form of If e and de, then Kp and Ki is used. The subscript d represents d-axis component and subscript q represents q-axis component. In this paper, these part illustrated in Fig. 5 are simulated in PSCAD/EMTDC and designed as new component using FORTRAN code. FLC is one of the most successful applications of fuzzy set theory. Its major features are the use of linguistic variables rather than numerical variables. The general structure of the FLC is shown in Fig. The FLC is composed of fuzzification, membership function, rule base, fuzzy inference and defuzzification. The fuzzification comprises the process of transforming crisp values into grades of membership for linguistic terms of fuzzy sets. The membership function is used to associate a grade to each linguistic term. For fuzzification the three variables of the FLC, the error (e), the change in error (de) and the gain output (Kp/Ki), have seven triangle membership functions for each. The basic fuzzy sets of membership functions for the variables are shown in the Figs. 7, 8 and 9. Fig. 7. The membership function for input fuzzy set (e and de) Kp and Ki Fig. 8. The membership function for Kp fuzzy set Fig. 9. The membership function for Ki fuzzy set TABLE II FUZZY CONTROL RULE BASE The change of error (de ) NB NM NS Z PS PM NB NB NB NM NM NS NS Z NM NB NM NM NS NS Z PB PS The error ( e) NS NM NM NS NS Z PS PS Z NM NS NS Z PS PS PM PS NS NS Z PS PS PM PM PM NS Z PS PS PM PM PB PB Z PS PS PM PM PB PB Fig. 5. Adaptive PI-Fuzzy Controller Fig.. Block Diagram of FLC All variables fuzzy subset for input and output are Negative Big (NB), Negative Medium (NM), Negative Small (NS), Zero (Z), Positive Small (PS), Positive Medium (PM) and Positive Big (PB) for all three variables. Table II shows the rule base for the FLC. The rules are set based upon the knowledge and working of the system. The gain values of Kp and Ki for PI controller of the current regulator are calculated for the changes in the input of the FLC according to the rule base. The number of rules can be set as desired. A rule in the rule base can be expressed in the form: If (e is NB) and (de is NB), then (Kp/Ki is NB). If (e is NB) and (de is NM), then (Kp/Ki is NB). If (e is NB) and (de is NS), then (Kp/Ki is NM). The rule base includes 9 rules, which are based upon the seven membership functions of the input variables to achieve the desired Kp and Ki.

5 5 In this work, Mamdani s max-min [1] method is used for inference mechanism. The center of gravity method [1] is used for defuzzification to obtain Kp and Ki, which is given by following equation: and 5 where, n is the total number of rules, μ i is the membership grade for the i-th rule, and C i is the coordinate corresponding to respective output or consequent membership function. V. SIMULATION RESULTS A symmetrical three-line-to-ground (3LG) fault at the transmission line as shown in Fig. 1 is considered as network disturbance. The fault occurs at.1 sec; the circuit breakers (CB) on the faulted line are opened at. sec, and at 1. sec the CBs are re-closed. In this transient analysis, the wind speeds for the wind generators are kept constant at the rated speed, assuming that the wind speed does not change dramatically within this small time duration. Three cases are considered in the simulation study to show the effectiveness of the proposed control strategy. In, simulation study is performed using the model system of Fig. 1 in which the adaptive PI-Fuzzy controller is applied to PMSG. In, simulation study is performed using the model system in which the conventional PI controller is applied, and in, simulation study is performed using another model system in which PMSG of Wind Farm 1 is replaced by IG (5 MVA). Simulations were performed by using PSCAD/ EMTDC. Figs. 1 and 11 show responses of reactive powers, from which it is seen that the grid side converter of Wind Farm 1 can provide necessary reactive power during the severe symmetrical 3LG fault in and. Therefore terminal voltages of the wind farms can return back to the rated value quickly in and as shown in Figs. 1 and 13. The rotor speeds can also become stable in and as shown in Figs. 1 and 15. From Figs. 1 through 15, it is seen that the wind farm voltage and rotor speed can be stabilized to the nominal value more quickly in the case of the adaptive PI- Reactive Power [MVar] Reactive Power [MVar] Fig. 1. Reactive power output of Wind Farm Fig. 11. Reactive power output of Wind Farm Fuzzy controller than the case of the PI controller. Figs. 1 and 17 show responses of the active power output of Wind Farm 1 and, respectively. From these results, it is clear that Terminal Voltage [pu] Terminal Voltage [pu] Rotor Speed [pu] Rotor Speed [pu] Active Power [MW] Active Power [MW] Fig. 1. Terminal voltage of Wind Farm Fig. 13. Terminal voltage of Wind Farm Fig. 1. Rotor speed of Wind Farm 1 Fig. 15. Rotor speed of Wind Farm Fig. 1. Active power output of Wind Farm 1 Fig. 17. Active power output of Wind Farm

6 VSWT-PMSG with the adaptive PI-Fuzzy controller can enhance the transient stability of total wind farm system significantly by controlling its active and reactive powers deli- performances vered to the grid as well as improve the system effectively. VI. CONCLUSION In this paper adaptive PI-Fuzzy controller is proposed in the two level back to back converter controllers of Variable Speed Permanent Magnet Wind Generator (PMSG) to enhance its transient performance as well as the performance of the neighboring wind farm composed of Fixed Speed Wind Generators. The controller combines fuzzy logic to classical PI controller to adjust online the PI gains. Stabilizing effect of the proposed PMSG system on the fixed speed wind generators is also investigated. The results show that the proposed adaptive PI- the transient Fuzzy controller is very effective in improving stability of overall wind farm system during a fault condition. VII. REFERENCES [1] M. H. Ali, T. Murata, J. Tamura, "Minimization of fluctuation of line power and terminal voltage of wind generator by fuzzy logic-controlled SMESS", International Review of Electrical Engineering, Vol. 1, No., pp , /1 [] J.L. Rodrı guez-amenedoa, S. Arnaltesa, M.A. Rodrı guezb, "Operation and coordinated control of fixed and variable speed wind farms," Renewable Energy, vol. 33, pp. 1, 8. [3] C.L. Souza et al, "Power System Transient Stability Analysis Including Synchronous and Induction Generator," in Proc. 1 IEEE Porto Power Tech, Vol., p.. [] H.F. Wang, F. Li, R.G. Cameron, "FACT control design based on power system non parametric model," in Proc IEE. -Gener. Trans. Distrib., Vol. 1, No. 5, pp [5] L. Gyugyi, "Unified power plow control concept for flexible ac transmission system," in Proc. 199 Inst. Elect. Eng. C, Vol No.. pp [] L. Gyugyi, "Dynamic compensation of AC Transmission line by solid state synchronous voltage source," IEEE Trans, Power Delivery, Vol. 9, No., pp , Apr.199. [7] Thomas Ackermann, Wind Power in Power System, UK: John Wiley & Sons, 5. p. 5. [8] Shuhui Li, Timothi A. Haskew, Ling Xu, "Conventional and novel control design for direct driven PMSG wind turbines," Journal of Electric Power System research, Vol. 8, pp , March. 1. [9] H. Polinder, S.W.H. de Haan, M.R. Dubois, J. Slootweg, "Basic operation principles and electrical conversion systems of wind turbines," in proc. Nordic Workshop on Power and Industrial Electronics, Paper 9, Trondheim, Norway. [1] G. Michalke, A.D. Hansen, T. Hartkopf, "Controll strategy of a variable speed wind turbine with multipole permanent magnet synchronous generator," in Proc. 7 European Wind Energy Conference and Exhibition, Milan (IT), May, 7. [11] S. M. Muyeen, Rion Takahashi, Toshiaki Murata, and Junji Tamura, "A variable speed wind turbine control strategy to meet wind farm grid code requirements, "IEEE Trans. Power System, vol. 5, pp , Feb. 1. [1] B. Ferdi, C. Benachaiba, S. Dib, R. Dehini, "Adaptive PI control of dynamic voltage restorer using fuzzy logic," Journal of Electrical Engineering: Theory and Application, Vol.1, pp , 1. [13] Siegfried Heier, Grid integration of wind energy conversion systems, John Wiley & Sons Ltd, p [1] S. M. Muyeen, J. Tamura, and T. Murata, Stability Augmentation of a Grid Connected Wind Farm, Green Energy and Technology, Springer- Verlag, 9, p. 73. VIII. BIOGRAPHIES Marwan Rosyadi (S 11) received Sarjana Teknik (equivalent to B.Sc. degree) from Adhi Tama Institute of Technology Surabaya and M.Eng degree from Sepuluh Nopember Institute of Technology, Indonesia, in and respectively, all in Electrical Engineering. Presently he is working towards his Ph.D Degree at the Kitami Institute of Technology, Hokkaido, Kitami, Japan. His research interests are power system stability and control including wind generator. S. M. Muyeen (M 8) received his B.Sc. Eng. Degree from Rajshahi University of Engineering and Technology (RUET), Bangladesh, formerly known as Rajshahi Institute of Technology, in and M. Sc. Eng. and Dr. Eng. Degrees from Kitami Institute of Technology, Japan, in 5 and 8 respec- and Electronic Engineering. tively, all in Electrical After completing his Ph.D. program he worked as a Postdoctoral Research Fellow under the versatile banner of Japan Society for the Promotion of Science (JSPS) from 8-1 at the Kitami Institute of Technology, Japan. Presently he is i working as Assistant Professor in Electrical Engineering department at the Petroleum Institute, in UAE. His research interests are power system stability and control, electrical machine, FACTS, energy storage system (ESS), Renewable Energy, and HVDC system. Rion Takahashi (M 7) received the B.Sc. Eng. and Dr. Eng. Degrees from Kitami Institute of Technology, Japan,, in 1998 and respectively, all in Electrical and Electronic Engineering. Now he is working as Associate Professor in Department of Electrical and Electronic Engineering, Kitami Institute of Technology. His major research interests include analysis of power system transient, FACTS and wind energy conversion system. Junji Tamura (M 87-SM 9) received his B. Sc. Eng. Degree from Muroran Institute of Technology, Japan, in 1979 and M.Sc. Eng. and Dr. Eng. degrees from Hokkaido University, Japan, in 1981 and 198 respectively, all in electrical engineering. In 198, he became a lecturer and in 198, an Associate Professor at the Kitami Institute of Technology, Japan, where he is i currently a Professor in the Department of Electrical and Electronic Engineering. His main fields of interest include analyses of power systems, electricall machine and wind energy conversion systems.