An Optimal Reactive Power Control Scheme in DFIG based Wind Power Generation with Reduction in Losses

An Optimal Reactive Power Control Scheme in DFIG based Wind Power Generation with Reduction in Losses

Sinchana.B.K, Student, 4th Sem, M.Tech (EPS),

Ghousia College of Engineering, Ramanagaram, Karnatakata, India

Dr. Mohd .Z.A. Anasari, Professor & HOD,

Ghousia College of Engineering, Ramanagaram, Karnatakata, India

Abstract- Power relations of doubly fed induction generator (DFIG) wind power generation system are analyzed. Based on this, a method is proposed to calculate the reactive power limit. The control strategy used at the wind generator level exploits a combination of pitch control and control of the static converters to adjust the rotor speed for the required operation points. This paper proposes an optimal reactive power dispatching strategy in order to minimize the total losses in a DFIG based wind farm, including the copper loss of the generators, the losses of converters, filters, transformers and the losses of cables. With the increasing penetration of wind turbines (WTs) grid utilities require extended reactive power supply capability not only during voltage dips but also in steady-state operation. A reactive power compensation strategy for the local user using DFIG wind farm is given. Simulation results are provided to verify the proposed theory.

Index Terms- Doubly fed induction generator, Wind turbines, Grid side converter, rotor side converter

1. INTRODUCTION

   Wind power has established itself as one of the most important renewable energy source over the past decades. With the priority status accorded to it in many countries, the share of wind power in relation to the overall installed capacity has increased significantly and this trend is in all likelihood set to continue. In some countries, the share of wind in relation to the overall installed capacity is already approaching the 50% mark. The increased prominence of wind in the generation mix inevitably leads to the question of its role in the provision of ancillary services, the most important of which being reactive power supply in support of grid voltage.

   The Doubly fed induction generator (DFIG) has many advantages such as good controllability, reduced power converter rating, etc., so it has been widely used in wind turbines. DFIG wind generation systems can provide or absorb reactive power from both the stator and the Grid Side Converter (GSC). Nowadays, the DFIG wind generation system is mostly required to operate at unity power factor status, which is a waste of its reactive power regulation ability, especially at low wind speed conditions. Due to the increased integration of wind energy into the power system, the wind farm are required more by the grid

   operator, and the reactive power control has become a major issue faced by wind farm operators.

   The most commonly used dispatch strategy is proportional dispatch, which spreads the required reactive power among all WTs proportional to their available reactive power. This method is easy to implement and is unlikely to exceed the reactive power limit of each WT. However, it does not consider active power loss in the WF. Another dispatch method proposed in considers the active power losses along the transmission cables and the transformers of WTs. However, this method does not consider the active power losses in the energy conversion system of WTs, which are responsible for a great part of the total loss in the WFs. Actually, the attempt to minimize active power losses along the transmission cables and the transformers may cause more losses in the energy conversion systems.

   An optimal dispatch strategy proposed in considered the losses from wind energy conversion systems, transformers and transmission cables, and found an optimal dispatch of reactive power for loss minimization. However, the strategy only used a simple WT control strategy and did not considered the influence of reactive power dispatch inside the Doubly Fed Induction Generator (DFIG) based wind energy generation systems. Since the DFIG energy generation system can regulate reactive power between the stator and the grid side converter (GSC), the reactive power flow inside the system will influence the losses of the system. Therefore the reactive power control method of the DFIG energy generation system should be studied. The most common method is to provide the reactive power only from the stator side by the rotor side converter (RSC). This method can reach good efficiency operating near Unity power factor, but the copper losses of the generator will increase significantly when the power factor increases. The second method is to regulate reactive power using both the RSC and the GSC to minimize copper losses. This method does not consider the losses from the converters and filters. It can reach a lower loss in certain choices of reactive power reference, but will increase the total loss in other cases. The method of splits the reactive power burden over the RSC and the GSC to reach minimum losses for the generator and converters. The splitting ratio is iteratively calculated, forming a set of look-up tables. In the control

   process, the controller should look up the tables to decide the optimal reactive power currents.

   In this paper, an optimal reactive power dispatch strategy is proposed for total loss minimization, which includes not

   Isq = Qs/Vs (4)

   where Qs is the reactive power of the stator. Deriving from (1), the rotor d-axis current and stator d-axis current can be calculated:

   only losses in the transmission cables and WT

   = – A – 1

   –

   (5)

   transformers, but also the losses inside wind energy

   generation system. The reactive power control of the WT uses optimal splitting strategy over the RSC and the GSC, which is implemented by solving an optimization problem

   = (6)

   that aims to minimize the total loss from the generator, the

   = B [ +

   + ] (7)

   converters and the filter. Consequently, reactive power

   dispatch between the WTs is integrated with the optimal

   where A = Rs/Xs, B = Xs/(2 + 2)

   reactive power control strategy of the WTs. The proposed

   strategy is then compared with traditional dispatch strategies in different cases.

   The copper losses in the DFIG can be calculated using:

   = (2 + 2 ) + (2 + 2 ) (8)

2. WIND FARM LOSS MODELS

   The WTs and the cables are the main devices that cause losses in a WF. The power losses of a WT consist of friction losses in the mechanical part, core losses and copper losses in the DFIG, losses in the converters and the filter, and the losses in the transformer of the turbine. The

   2. Loss Model of Converters and the Filter

   The losses in the converter, which consists of transistors and reverse diodes, can be divided into switching losses and conducting losses. The losses in a converter can be expressed as

   friction losses and core losses can be considered constant

   =

   + 2 (9)

   under a certain operation point; therefore they are not

   1

   1

   considered in this paper. In the following paragraphs, the loss models of each component are derived.

   1. Loss Model of DFIG

   where Irms is the rms value of the sinusoidal current at the converter ac terminal, and al and bl are the power module constants and can be expressed as

   1= 6 2 ( + + ) (10)

   For a DFIG operating in a stator voltage orieted reference frame, the steady-state voltage equations are as follows. All

   .

   .

   variables in the equations are in per unit (pu) system.

   1= 3 (11)

   0

   0

   0

   where VIGBT is the voltage across the collector and emitter of the IGBT, EON + EOFF is the total turn-on and turn-off

   =

   '

   (1)

   losses of the IGBTs, IC,nom is the nominal collector

   [ ]

   0 [ 0

   ]

   [ ]

   current of the IGBT, fswis the switching frequency, Err is the turnoff (reverse recovery) loss of the diodes, rIGBT is

   Where stator inductance Xs equals Xls + Xm, rotor inductance equals + Xm, Xls is stator leakage inductance, Xm is mutual inductance, is rotor leakage

   inductance, s is rotor slip. The subscripts are s, r and g for stator, rotor and grid converter circuits; l and m for leakage

   the lead resistance of the IGBT.

   The current flows through Rotor Side Converter (RSC) and GSC can be calculated as:

   = (2 + 2 )

   and mutual inductances; d and q for direct and quadrature

   axes. The superscript is used for rotor value referred to the

   = (2 + 2 ) (12)

   stator. At a fixed wind speed, the rotor d-axis current is

   constant, and can be calculated as

   The grid side converter d-axis current Igd can be

   = –

   (2)

   calculated:

   = (3)

   where Pmec is the power extracted from the wind, r is the angular frequency of the voltages and currents of the rotor windings, s is the angular frequency of the voltages and

   = ( + ) / (13)

   The grid side converter q-axis current Igq can be calculated:

   currents of the stator windings, u is the turns ratio. The stator q-axis current can be calculated as:

   =

   (14)

   = (15)

   where Qg is the reactive power provided by the GSC and QWT is the total reactive power from/to the WT. With the grid side converter currents, the loss in the grid side filter can be calculated using:

   = (2 + 2 ) (16)

   paper. The formulas of both strategies are presented in the section.

   Strategy 1(S1): Traditional Proportional Dispatching Strategy

   The strategy dispatches the required reactive power proportionally among all operative generators based on

   their available reactive power which can be expressed in

   So, the total loss from a WT is :

   = + + + (17)

   =

   =1

   (21)

   where is the reactive power that wind turbine i must

   1. Loss Model of Transformers

      generate;

      the maximum reactive power that wind

      The active power loss in transformers can be calculated as

      = +2 (18)

      where is the load ratio, P0 is the no-load loss, and Pk is the load loss.

   2. Loss Model of Cables

   Consider the cable connecting the two buses i and j in Fig. 1, where y and I mean the admittance and current of each cable, and V means the voltage on each bus. The cable current, Iij , measured at bus i and j defined positive in the direction i j is given by

   Iij = Il + Ii0 = yij (Vi Vj) + yi0Vi. (19)

   turbine i can generate in one specific moment; n is the number of wind turbines and is the total reactive power required by the wind farm operator. The wind turbines are controlled by setting reactive power of the GSC equals to zero, so the reactive power is generated by the RSC. Therefore the maximum reactive power that can be provided is limited by the maximum current of the RSC. The total reactive power required by the wind farm

   operator of Strategy 1 must include the reactive power requirement of the Point of Common Coupling (PCC) and the reactive power loss on the cables.

   Strategy 2(S2): Loss minimization dispatching strategy

   The proposed strategy aims to minimize the total losses of the whole system in the wind farm. The objective function is:

   Similarly, the cable current Iji is given by

   min

   +

   (22)

   =1

   =1

   Iji = Il + Ij0 = yij (Vj Vi) + yj0Vj . (20)

   The power loss in cable ij is the algebraic sum of the complex powers Sij from bus i and j and Sji from bus j and i,

   where is the active power loss of wind turbine i , n is the total number of wind turbines,

   is the active power loss of cable k, m is the total number of cables.

   The main constraint is the maximum available reactive power constraint, which is limited by the maximum current of the RSC, so it can be expressed as:

   (23)

   Where can be calculated using (12), is the

   Fig. 1: The model circuit of a Cable

   rated collector current of RSC.

   The losses of cables in a wind farm are calculated after power flow computation using Newton-Raphson method in

   this paper.

   Another constraint is the voltage limit at each bus, as expressed as follows:

   (24)

3. OPTIMIZATION PROBLEM FORMULATION

   Proportional dispatching method is a traditional strategy for reactive power dispatching. It is compared with the

   where k is the number of the bus. Also, the output reactive power should follow reactive power requirement at PCC. So the constraint can be expressed as:

   proposed loss minimization dispatching strategy in this

   –

   = (25)

   =1

   =1

   where is the reactive power set point of wind turbine

   i, is the reactive power loss on cable k , is the

   the balances the system. Filter produces constant reactive power.

   reactive power requirement of the PCC. So, Strategy 2 can be modelled as an optimization problem with the objective

   1. and the constraints (23), (24), (25).

      Strategy 3(S3): Optimization Methods

      Sequential quadratic programming method is used in the paper to solve the optimization problem under constraints. This method makes a lot of iterations in order to find the optimization results under the constraints. At all iteration, an approximation is made of the Hessian matrix of the Lag- rangian function using a Quasi-Newton updating method. This is then used to generate a quadratic programming sub- problem whose solution is used to form a search direction for a line search procedure.

4. SIMULATION MOODEL OF DFIG

   Fig 2: Simulation model of DFIG system connected to Grid

   Fig 3: The DFIG system which is connected to back-to-back converters

   A schematic diagram of a wind energy conversion system of DFIG is shown in Fig. 2 and Fig 3. The stator of the DFIG is directly connected to the grid through buses to supply the active and reactive power. In the rotor circuit, there are back-to-back converters (grid side converter, rotor side converter, filters) inserted between rotor and gird. The grid converter is used to regulate DC bus voltage speed. This simulation model is done by using MATLAB Simulink software.

   The doubly fed induction generator provides reactive power to grid and initially reactive power reference is taken as zero. When mre reactive power is required to the consumers or when the inductive loads are high, then the DFIG generates reactive power. Similarly, When the inductive loads are low, the DFIG absorbs reactive power

5. SIMULATION RESULTS

The simulation results of the doubly fed induction generator are shown in the below fig 4 and fig 5. With the theoretically basics of loss models and with the help of simulation results, the least power loss is produced by the machine than compare to rotor side converter, grid side converter and filters. Hence the machine with fed the reactive power to the gird for further utilization by consumers. This is shown is the graphs below.

Fig 4: the power loss resuts of RSC, GSC and filter

Fig 5: the power the results of the machine.

The figure 6 shows the real and reactive power feeding to the gird. The results are in per unit value. The graph shown below is generating reactive power with constant real power. At starting the waves are more oscillating because the machine needs few second to change its operating mode.

Fig 6: Real and reactive power results

CONCLUSION

This paper introduces an optimal reactive power dispatching strategy to minimize the total losses in a wind farm. On the basis of the loss model, an optimization problem is built considering the operating constraints. Proposed strategy is compared with traditional proportional dispatching strategy. It can be concluded from the simulation results that the proposed strategy is effective in minimizing operating loss of the whole system. The DFIG- based WT systems can adjust active and reactive power output independently. The implementation of this optimized strategy requires a modification on the WT control level, i.e., each WT should be able to follow two reactive power references by controlling the RSC and the GSC.

REFERENCES

1. N. R. Ullah, K. Bhattacharya, and T. Thiringer, Wind farms as reactive power ancillary service providersTechnical and economic issues, IEEE Trans. Energy Convers., vol. 24, no. 3, pp. 661672, Sep. 2009.

2. A. Tapia, G. Tapia, and J. X. Ostolaza, Reactive power control ofwind farms for voltage control applications, Renew. Energy, vol. 29, pp. 377392, 2004.

3. Y. Lang, X. Zhang, D. Xu, H. F. Ma, and S. R. Hadianmrei, Reactive power analysis and control of doubly fed induction generator wind farm, Proc. CSSE, vol. 27, no. 9, Mar. 2007, pp. 7782.

4. J. R. Saenz, A. Tapia, G. Tapia, F. Jurado, X. Ostolaza, and I. Zubia, Reactive power control of a wind farm through different control algorithms, in Proc. 4th IEEE Int. Conf. Power Electron. Drive Syst., Denpasar, Indonesia, 2001, pp. 203207.

5. R. G. De Almeida, E. D. Castronuovo, and J. A. Peas Lopes, Optimum generation control in wind parks when carrying out system operator requests, IEEE Trans. Power Syst., vol. 21, no. 2, pp. 718725, May 2006.

6. B. Kanna and S. N. Singh, Towards reactive power dispatch within a wind farm using hybrid PSO, Int. J. Elect. Power Energy Syst., vol. 69, pp. 232240, 2015.

7. M. Martinez-Rojas et al., Reactive power dispatch in wind farms using particle swarm optimization technique and feasible solutions search, Appl. Energy, vol. 88, no. 12, pp. 46784686, 2011.

8. B. Zhang, W. Hu, P. Hou, and Z. Chen, Reactive power dispatch for loss minimization of a doubly fed induction generator based wind farm, in Proc. IEEE 17th Int. Conf. Elect. Mach. Syst. (ICEMS14), Hangzhou, China, 2014, pp. 13731378.

9. M. Kayikci and J. V. Milanovic, Reactive power control strategies for DFIG-based plants, IEEE Trans. Energy Convers., vol. 22, no. 2, pp. 389396, Jun. 2007.

10. F. Iov, F. Blaabjerg, and A. D. Hansen, Analysis of a variable- speed wind energy conversion scheme with doubly-fed induction generator, Int. J. Electron., vol. 90, nos. 1112, pp. 779794, 2003.

11. R. Pena, J. C. Clare, and G. M. Asher, Doubly fed induction generator using back-to-back PWM converters and its application to variable-speed wind-energy generation, Proc. IEE Elect. Power Appl., vol. 143, no. 3, pp. 231241, Sep. 1996.

12. K. Chen, P. Delarue, A. Bouscayrol, P. Vidal, and M. Pietrzak- David, Minimum copper loss and power distribution control strategies of double-inverter-fed wound-rotor induction machines using energetic macroscopic representation, IEEE Trans. Energy Convers., vol. 25, no. 3, pp. 642651, Sep. 2010.

13. H. Li, Z. Chen, and J. K. Pedersen, Optimal power control strategy of maximizing wind energy tracking and conversion for VSCF doubly fed induction generator system, in Proc. IEEE 5th Int. Power Electron. Motion Control Conf., Shanghai, China, 2006, vol. 6, pp. 16.

14. Y. Tang and L. Xu, A flexible active and reactive power control strategy for a variable speed constant frequency generating system, IEEE Trans. Power Electron., vol. 10, no. 4, pp. 472 478, Jul. 1995.

15. K. Krajangpan, W. Sdara, and B. Neemmanee, Control strategies for maximum active power and minimum copper loss of doubly fed induction generator in wind turbine system, in Proc. Int. Conf. Power Syst. Technol., Hangzhou, China, 2010, pp. 17.

16. B. Rabelo andW. Hofmann, Control of an optimized power flow in wind power plants with doubly fed induction generators, in Proc. IEEE 34th Annu. Power Electron. Spec. Conf., 2003, pp. 15631568.

17. B. Rabelo, W. Hofmann, and L. Pinheiro, Loss reduction methods for doubly-fed induction generator drives for wind turbines, in Proc. Int. Symp. Power Electron. Elect. Drives Autom. Motion, Taormina, Italy, 2006, pp. 12171222.

18. G. Abad et al., Dynamic modeling of the doubly fed induction machine, in Doubly Fed Induction Machine: Modeling and Control for Wind Energy Generation Applications. Hoboken, NJ, USA: Wiley, 2011, pp. 231234.

19. A. Petersson, Analysis, modeling and control of doubly-fed induction generators for wind turbines, Ph.D. dissertation, Dept. Energy Envir., Chalmers Univ. Tech., Goteborg, Sweden, 2005.