Published on

IJERA (International journal of Engineering Research and Applications) is International online, ... peer reviewed journal. For more detail or submit your article, please visit www.ijera.com

Published in: Technology
  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide


  1. 1. Jagannath Ch yadav .B, KVR Swathi / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue 5, September- October 2012, pp.102-106 Simulation and Control of A Doubly Fed Induction Generator (DFIG) Using Artificial Neural Networks Jagannath Ch yadav .B1 KVR Swathi2 1 Department of Electrical and Electronics Engineering, ANITS College, Visakhapatnam, Andhra Pradesh, India. 2 Assistant Professor, Department of Electrical and Electronics Engineering, ANITS College,Visakhapatnam, Andhra Pradesh, India.ABSTRACT The paper describes the design of a Moreover, the capacity of the excitation powerdoubly fed induction generator (DFIG), using supply is only 25%~30% of the whole power unit.back-to-back PWM voltage-source converters in This would save a large expense to a certain extent.the rotor circuit. A vector-control scheme for the Because of these advantages, DFIG is one of thesupply-side PWM converter results in main generators in the wind farms. The structure ofindependent control of active and reactive power the DFIG system is as shown in fig1[1]. The winddrawn from the supply.Slip ring induction turbine is connected with the generator rotor by amachine in the variable speed wind turbine gearbox. The stator of the generator could parallelpopularly known as double fed induction with the electricity grid through a contractor.generator is mostly used in wind power Induction generator is coupled with a windgeneration. The main reasons for the popularity of mill offers an ideal solution. Wind energy is availablethe doubly fed wind induction generators in abundance in our environment. When comparedconnected to the power network is their ability to with the conventional sources of energy, wind energysupply power at constant voltage and frequency is clean, efficient, and sustainable form of energy.while the rotor speed varies and motor converter When the cost of supplying electricity to remotehandles fraction of stator power. The main goal of locations is expensive, wind energy provides a costmy project is to design doubly fed induction effective alternative. So to convert this wind energygenerator (DFIG) & to control the rotor and into electrical energy, an induction generator willstator voltages by injecting the proper rotor offer an ideal solution.voltage to the DFIG derived from PI and ANNcontroller so as to maintain the constant terminal II. WIND TURBINE WITH A DOUBLYvoltage.The results also verify the simulation FED INDUCTION GENERATOR:model can meet the requirement of the VSCF Wind turbines use a doubly-fed inductionwind energy conversion system. generator (DFIG) consisting of a wound rotor induction generator and an AC/DC/AC IGBT-basedKEYWORDS - DFIG, Stationary Reference frame, PWM converter. The stator winding is connectedDynamic Model, Artificial Neural Networks. directly to the 50 Hz grid while the rotor is fed at variable frequency through the AC/DC/ACI. INTRODUCTION: converter[3]. It is a 4-quadrant converter which can Wind power is a clean, non-polluting and realize bidirectional flowing of energy. On the onerenewable energy. There are many advantages of hand, the rotor can absorb energy from power gridwind power conversion system compared with the and on the other hand, it can feed the energy back tothermal power generation unit and nuclear power the power grid under certain conditions. 4-quadrantunit[1]. For example, wind power conversion system means the excitation power supply can operate oncan reduce the emissions of CO2 and other harmful positive resistive, pure capacitive, negative resistivegases. A 1-MW wind power generator reduces 2000 and pure inductive the four typical state.tons CO2,10 tons SO2 and 6 tons of NO2 emissionsto the atmosphere each year [3]. Therefore, windenergy is always known as “blue sky anthracite”.Among the renewable energy conversion systems,wind power has the most commercial prospects.Especially in the last few years due to the rapiddevelopment of wind power industry, the world windpower capacity increased by a linear trend. The DFIGcan achieve VSCF power generation and decoupled Fig 1. Wind turbine with a doubly-fed inductionstator active and reactive power control due to its AC generator (DFIG)variable-frequency excitation power supply. 102 | P a g e
  2. 2. Jagannath Ch yadav .B, KVR Swathi / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue 5, September- October 2012, pp.102-1062.1. Operation of DFIG: understand vector control principle, a good The stator is directly connected to the AC understanding of d-q model is mandatory.mains, whilst the wound rotor is fed from the Power The transformation equation from a-b-c to this d-q-oElectronics Converter via slip rings to allow DIFG to reference frame is given byoperate at a variety of speeds in response to changing fqdo=Ks*fabcs (5)  wind speed. Indeed, the basic concept is to interposea frequency converter between the variable frequency where, (f qd 0 s )T  f qs f ds f0s ,induction generator and fixed frequency grid[4],[5]. (f abcs )T   f as f cs ,The DC capacitor linking stator- and rotor-sideconverters allows the storage of power from f bsinduction generator for further generation. To achievefull control of grid current, the DC-link voltage must  2 2  cos  cos(  3 ) cos(  ) 3 be boosted to a level higher than the amplitude of 2 2 2 grid line-to-line voltageThe slip power can flow in K s   sin  sin(  ) sin(  ),both directions, i.e. to the rotor from the supply and 3 3 3   1 1 1 from supply to the rotor and hence the speed of the  2   2 2 machine can be controlled from either rotor- orstator-side converter in both super and sub-synchronous speed ranges. At the synchronous speed, Where the variable f can be the phaseslip power is taken from supply to excite the rotor voltages, current, or flux linkages of the machine.windings and in this case machine behaves as a The transformation angle  r between the q- axis ofsynchronous machine.The mechanical power and the the reference frame rotating at a speed of w and the a-stator electric power output are computed as follows axis of the stationary stator winding may bePr=Tm*ωr (1) expressed asPs=Tem*ωs (2) t (6)For a loss less generator the mechanical equation is     (t )dt  (0).J (dωr/dt) =Tm-Tem (3) 0In steady-state at fixed speed for a loss less generator 3.1 Torque Equations:Tm=Tem and Pm=Ps+Pr (4) The sum of the instantaneous input power toAnd it follows that: all six windings of the stator and rotor is given by:Pr=Pm-Ps=Tm ωr- Tem*ωs =-S Ps Pin=Vas Ias + Vbs Ibs+ Vcs Ics + Var Iar+ Vbr Ibr+Where Vbr Ibr (7)S= (ωs- ωr)/ ωs is defines as the slip of the generator. Using stator and rotor voltages to substitute for the voltages on the right hand side of (4.8), we2.2. Back-to-Back AC/DC/AC Converter obtain three kindsof terms: i2r, idψ/dt and ωψi.(i2r)Modeling: terms are the copper losses. The electromagnetic Mathematical modeling of converter system torque developed by the machine is given by the sumis realized by using various types of models, which of the (ω.ψi) terms divide by mechanical speed, thatcan be broadly divided into two groups: mathematical is:functional models and Mathematical physical models Tem=(3/2) (p/2 ωr)[ ω(ψds iqs-ψqs ids)+( ω- ωr)((either equation-oriented or graphic-oriented, where ψdr iqr-ψqr idr)] (8)graphic-oriented approach is actually based on the Using the flux linkage relationships, Tem can also besame differential equations)[6]. expressed as follows: Tem=(3/2) (p/2 ωr)[ ω(ψds iqs-ψqs ids)+( ω- ωr)(III.MATHEMATICAL REPRESENTATION ψdr iqr-ψqr idr)] (9)OF DFIG: Using the flux linkage linkage relationships, one can Assuming that the three-phase power grid show thatvoltages are balance and the three-phase circuits are Tem=(3/2) (p/2)[( ψqr idr-ψdr iqr)]symmetrical, the sum of the three-phase currents in =(3/2) (p/2)[( ψds iqs-ψqs ids)]the three-phase non-median line system is zero. =(3/2) (p/2)Lm[( idr iqs- iqr ids)] (10) An induction motor can be looked on as a One can rearrange the torque equations by insertingtransformer with a rotating secondary, where the the inserting the speed voltage terms given below:coupling coefficients between the stator and rotor Eqs= ωψds Eds= -ωψqsphases change continuously with the change of rotor Eqr= (ω- ωr)ψdr Edr= -(ω- ωr)ψqr.position[7]. The machine model can be described bydifferential equations with time varying mutual 3.1.1 Induction Machine Equations In Stationaryinductances, but such a model tends to be very Reference Frame:complex, such as vector control, based on the Stator circuit equations:dynamic d-q model of the machine. Therefore to Vqss=d/dt (ψqss)+rs iqss (11) 103 | P a g e
  3. 3. Jagannath Ch yadav .B, KVR Swathi / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue 5, September- October 2012, pp.102-106Vdss=d/dt (ψdss)+rs idss (12) The electromagnetic torque developed is Te=2Hd/dt (ωm) + Bm ωm + Tl (17)Rotor circuit equations: Where Te=Tg and Tshaft=Tl)Vqrs=d/dt (ψqrs)+rr iqrs (13) By neglecting the torque due to friction (Bm ωm) s s s Te-Tl=2Hd/dt(ωm) (18)Vdr =d/dt (ψdr )+rr idr (14) From above equation, the rotor speed (ωm) isFlux linkage equations: ωm = (Te-Tl)/ (2Hg) dt = (0.5/Hg) (Te-Tl) dt (19)ψqss=Lls iqss+Lm(iqss+iqrs)=(Lls+Lm)iqss+Lm similarly the turbine speed isiqr =Ls iqs +Lm iqrs s s (15) ωt = (Tl-Tw)/ (2Hw) dtψqrs=Llr iqrs+Lm(iqss+iqrs)=(Llr+Lm)iqrs+Lm = (0.5/Hw) (Tl-Tw) dt (20)iqs =Ls iqr +Lm iqss s s from the above equations, we have Tl= Km (θm- θt) = Km (ωm- ωt) dtψdss=Lls idss+Lm(idss+idrs)=(Lls+Lm)idss+Lm where θ= ω dtidr =Ls ids +Lm idrs s s In the doubly fed induction generator isψdrs=Llr idrs+Lm(idss+idrs)=(Llr+Lm)idrs+Lm runned at a specified speed with the statorids =Lr idr +Lm idss s s disconnected from the grid. The rotor is suddenly excited with the slip frequency voltages derived from3.1.2. d-q Torque Equations: voltage regulator so as to produce commanded openTem= (3/2) (p/2) [( ψqr idr-ψdr iqr)] circuit stator terminal voltage. The specified = (3/2) (p/2) [( ψds iqs-ψqs ids)] operating conditions and final values of the variables = (3/2) (p/2)Lm[( idr iqs- iqr ids)] (16) reached in the steady state are all saved in the workspace to serve as initial conditions in aIV.SIMULINK IMPLIMENTATION OF subsequent simulation.DFIG: Based on the control strategy discussedabove, Fig.6 shows the rotor side converter controlstrategy with current regulated PWM. A current-regulated PWM voltage source inverter provides fieldoriented currents iqr and idr to the rotor circuit,controlling active power and statorreactive power,respectively. Active power command is given by theturbine optimal torque speed profile and reactivepower command is calculated to minimize themachine copper losses. Overall active powergenerated is directly related to the torque, asindicated by (7) and (8). In the d-q reference frame asdetermined by the machine stator flux, its currents iqr Fig.2: DFIG with PI CONTROLLERand idr are also field oriented, controlling Active andReactive power of grid (Ps and Qs) respectively. Simulink is a software package formodeling, simulating, and analyzing dynamicalsystems. It supports linear and nonlinear systems,modeled in continuous time, sampled time, or ahybrid of the two. Systems can also be multirate, i.e.,have different parts that are sampled or updated atdifferent rates. This is a far cry from previoussimulation packages that require you to formulatedifferential equations and difference equations in alanguage or program. Simulink includes acomprehensive block library of sinks, sources, linear Fig.3: Dynamic Model of Induction Machine inand nonlinear components, and connectors. You can Arbitrary Reference Framealso customize and create your own blocks. For The rotor-side converter is used to controlinformation on creating your own blocks, see the the wind turbine output power and the voltageseparate Writing S-Functions guide. measured at the grid terminals. The power is controlled in order to follow a pre-defined power-4.1. Simulink Implementation Of Mechanical speed characteristic, named tracking characteristic.System This characteristic is illustrated by the ABCD curve 104 | P a g e
  4. 4. Jagannath Ch yadav .B, KVR Swathi / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue 5, September- October 2012, pp.102-106superimposed to the mechanical power characteristicsof the turbine obtained at different wind speeds.Fig.4: Simulink Diagram for Rotor SideConverter Fig.7 DFIG with ANNFig.5: Simulink Diagram for Stator SideConverterArtificial Neural Networks Numerous advances have been made indeveloping intelligent systems, some inspired bybiological neural networks. Researchers from manyscientific disciplines are designing artificial neuralnetworks to solve a variety of problems in patternrecognition, prediction, optimization, associative Fig.8: Stator Output Voltage and Current usingmemory, and control. Conventional approaches have PI controllerbeen proposed for solving these problems. Althoughsuccessful applications can be found in certain well-constrained environments, none is flexible enough toperform well outside its domain[8]. ANNs provideexciting alternatives, and many applications couldbenefit from using them. This article is for thosereaders with little or no knowledge of ANNs to helpthem understand the other articles in this issue ofComputer. Fig.9: Stator Output Voltage and Current using Artificial Neural NetworksFig. 6 Simple Artificial Neuron VII. CONCLUSIONS: The long course of evolution has given the The modeling and simulation of a DFIG was carriedhuman brain many desirable characteristics not out with vector control of the parameters in the statorpresent Invon Neumann or modern parallel voltage DQ reference frame. The DFIG was modeledcomputers. These include massive in both Matlab software packages. The behavior ofparallelism,distributed representation and the models was examined by injecting rotor voltagecomputation, learning ability,Generalization ability. at a constant torque. An attempt was also made to derive the existing controllers adopted in the rotor The above figure shows three phase open side converter to control the output Voltage and thecircuit voltages ea, eb, ec which are displaced by 120 current of the DFIG. The PI controller and ANN areelectrical degrees apart. Hence from this we can say used for grid control. As compared the responsethat power is generated from doubly fed induction between two controllers ANN gives the better results.generator. 105 | P a g e
  5. 5. Jagannath Ch yadav .B, KVR Swathi / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue 5, September- October 2012, pp.102-106REFERENCES: [1] A.Dendouga, R. Abdessemed, M. L. Bendaas and A. Chaiba LEB-Research Laboratory, Department of Electrical engineering, Batna University, Algeria (2007)“Decoupled Active and Reactive Power Control of a Doubly-Fed Induction Generator (DFIG)”, proceedings of the 15 th Mediterranean conference on control & Automation, July 27-29, Athens,-Greece. [2] Fan Xiaoxu, Lv Yuegang, Bai Yan, Xu Daping, “Modeling and Simulation of the Variable-frequency Excitation Power Supply Served for the Wind Power DFIG(2008)”. [3] Chitti Babu B, K.B.Mohanty, C. Poongothai (2009), “Steady State Analysis and Control of Wind Turbine Driven Double-Output Induction Generator”,3rd International Conference on Power Electronics Systems and Applications. [4] S. K Salman and Babak Badrzadeh School of Engineering, The Robert Gordon University, IEEE paper on, “New Approach for modeling of Doubly-Fed Induction Generator (DFIG) for grid-connection studies”. [5] S. Muller, M. Deicke and R.W. De Doncker (2002), "Doubly fed induction generator systems for wind turbine", IEEE Industry Applications Magazine, Vol.8, No. 3, pp. 26-33. [6 Peterson A. Analysis (2003), “Modeling and control of Doubly-Fed Induction Generators for wind turbines”. Licentiate thesis; Chalmers University, Gutenberg, Sweden. [7] Ong CM (1998) “Dynamic Simulation of electric machinery using MATLAB/SIMULINK”, Prentice Hall. [8] LiMin Fu, “Neural Networks in Computer Intelligence”, McGraw-Hill Inc., pp. 153- 264, 1994.. 106 | P a g e