Transient Stability Improvement Analysis of DFIG Based Variable Speed Wind Generator Using IGBT- Bridge-Type Fault Current Limiter

: Due to its many advantages such as the improved power quality, high energy efficiency and controllability, etc. the variable speed wind turbine using a doubly fed induction generator (DFIG) is becoming a popular concept and thus the modeling of the DFIG based wind turbine and improvement in the transient fault conditions is an important consideration. In this paper, Transient stability improvement has been done by using Superfluous Fault Current Limiter [SFCL]. A new design of SFCL IGBT- Bridge-type SFCL shunted with a variable resistor Rsh has been proposed. Rsh is modeled to decrease the terminal Voltage deviation to minimum level by reducing the amount of current at the bus terminal. Comparison of Voltage deviation and current deviation with the resistive type SFCL and IGBT- bridge-type SFCL show considerable decrease in both quantities by using IGBT- bridge-type SFCL. The values of Voltage deviation at the bus terminal is 8.223 e-8 % for proposed SFCL which is less than the resistive type SFCL that is, 14.4 e -8 %. The huge Voltage sag has been considerably reduced by reduction of high level of current to 0.0004401 % in IGBT-bridge-Type SFCL from 0.0004624% in resistive type SFCL. Thus proposed SFCL has caused significant improvement in transient stability keeping the deviation in active and reactive power during faults to minimum level.


INTRODUCTION
Renewable energies have the potential to reduce pollution, slow global warming and create new industries and jobs. There are huge economic opportunities for countries that invent, produce and export clean energy technologies. The use of renewable energies has become a global research and development problem. Wind power is one of the fastest growing renewable technologies and has the potential to meet a significant portion of our electricity needs. It is now also considered an energy source that allows electricity to be produced with minimal environmental disturbance. In recent years, dual-fuel induction generators (DFIGs) have been one of the most used large gridconnected wind turbines. They are becoming more and more acceptable due to their suitability for variable speed wind turbines and are more attractive than fixed speed systems due to their efficient power generation, better power quality and performance. Performance of electronic devices, which carry only a fraction of the total power. (20-30%). J. Wang et al. [2] This article uses hierarchical grouping methods to classify the wind farm, performs hierarchical cluster analysis for all wind turbines, and divides it into multiple clusters using the transient Voltage properties of each wind turbine as the cluster target. A concentrated process, which uses the actual performance characteristics of the cluster as a benchmark, manages the identification of the parameters. The first WTG models, which are mainly based on electromagnetic transient models and contain many power electronic components, are complex and have low convergence and low computation speed.
Y. Lei et al. [3] in this paper, a simple DFIG wind turbine model is developed in which the converter is simulated as a controlled Voltage source and the rotor current is adjusted to satisfy the active and reactive power generation command. The performance and accuracy of the model were also compared with the detailed model developed by DIgSILENT.
Z. Zhao et al. [4] In this article, taking into account the problem of wind speed difference due to complex terrain and irregularly arranged wind farms in large wind farms, measurement data on a long-term scale is selected as a cluster-dependent index. The simulation results show that the equivalent dynamic model of the created wind farm can accurately reflect the dynamic nature at the common coupling point (PCC In modeling of turbine model, we consider the most common relationship between the wind speed and the extracted mechanical power P wind by the wind turbine, which can be defined as follows P wind = 0.5 ρ C p (λ,β) a w 3 Watts where ρ is the air density, s w the wind speed, a w is the area covered by the rotor of the wind turbine, a w = πr 2 where r= radius of the blade, Cp is the power conversion coefficient which is the function of both tip speed ratio, λ, and blade pitch angle, β. By controlling the blade pitch angle (β), the wind turbines can extract more wind energy within the wide range area of wind speed. The tip speed ratio is defined as follows: λ= ( / s w ) * r where ω is the rotational mechanical speed (rad/s). For the modeling of wind turbine, Cp (,-) can be calculated as: Where: The coefficients from c1 to c6 are considered as follows: C 1 = 0.5176, C 2 = 116, C 3 = 0.4, C 4 = 5, C 5 = 21, AND C 6 = −0.0068.
The basic diagram of DFIG with its two converters (RSC and grid-side converter (GSC)) is shown in Figure 3 where in equations (5) and (6), is the rotor-side Voltage; I r is the current of the rotor; is the rotor winding resistance; is the flux vector for rotor; L m is the magnetizing inductance; is is the current of the stator; L r is the rotor windings selfinductance; and finally, r and s are the rotor and stator subscripts, respectively.
The rotor self-inductance (L ) r can be defined.

= +
where Llr is the leakage inductance for rotor.
In RSC, the idr and irq are processed with the help of PI controllers, and give the Voltages of v′rd and v′qr , respectively,and it is defined in equations (8)  In order to ensure the decent tracking of the dq axis current, in RSC controller, the compensation terms are added to ΄ and ΄ to obtain the reference Voltages ( * and * ), which are defined in the below equations: where in equations (10) and (11), is the slip, and it is defined as = -.
The pulses are generated through the PWM generator after calculation of reference for the insulated-gate bipolar transistor (IGBT). They are then fed to the control bridges on the rotor side in Mat lab/ Simulink as gate pulses.

C. Grid Side Controlling System:
The main purpose of the GSC is to keep fixed DC link Voltage and to keep a constant power factor. To achieve this goal the vector control method is used. In DFIG, the GSC ensures balanced power energy on the both sides of the DC link capacitor by maintaining the DC link Voltage.
In order to design the GSC, two series PI controllers are used in this research work. The gain parameter values of the GSC controller are K p = 0.84 and K i = 5. The GSC controller is made up of a universal bridge converter in MATLAB/ SIMULINK with snubber resistance 1e3 ohms and uses IGBT as the power electronic device. The IGBT uses gate pulses as controlling signals.
It uses the DC link Voltage v dc and the reactive power (Q ) s from the rotor line as inputs and sends the desired signal with the processing of the PI controller and carrier frequency of the GSC controller.  (11) and (12), is the stator side Voltage; is the current of the rotor; is the stator winding resistance; is the flux vector for stator; is the current of the stator; is the magnetizing inductance; is the stator Windings self-inductance; and finally, r and s are the rotor and stator subscripts, respectively.
The stator self-inductance can be defined as: is the leakage inductance for stator In GSC, the d-axis reference component is associated with the grid Voltage angular position α s . As the grid Voltage amplitude is constant, is zero and is constant. The active power and reactive power will be proportional to the and , respectively.
The converter active power flow ( )and the reactive power ( ) are defined in equations (15) and (16), respectively,which demonstrated the real and reactive powers from the GSC, which are controlled by and current components, respectively where in equations (15) and (16), d, q, and con represent the axis component subscripts and converter value subscript, respectively.
In order to appreciate the decoupled control, the similar compensation also introduced in equations (17)  After calculating the reference Voltages, inverse-park transformation is used in order to give the appropriate three phase Voltage (v c abc * ) to the final PWM pulse generator for the GSC converter. Due to this, the current (I SFCL ) flowing through the superconducting coil is unidirectional. This helps minimise the loss across the coil, L SC . Although there are some power losses across the rectifier diodes, it is reported that using the DC resistive SFCL provides better system efficiency, even when considering these losses The DC resistive SFCL is designed in such a way that the magnitude of quench resistance varies exponentially from 0.0 to 2.0 pu. Generally in DC resistive SFCL, the resistance value of Rc is kept zero during the normal operation. During a fault, a value of quenched resistance (R sh ) = 0 65 pu is considered in this article to compare the transient stability performances among the series compensators. The main advantage of using the DC resistive SFCL is that no additional controller is required to change from the non-superconducting state to superconducting states. When V B1 is slightly greater than the V ref , the controller forces the IGBT gate signal move to a higher state, which makes the IGBT switch turn on. As the IGBT switch is on, the shunt path is withdrawn from the operation, and normal operation of the system returns and continues.

E. Proposed Design Of IGBT-Bridge-Type SFCL
A novel concept of an SFCL acting as a rectifier-type SFCL in normal conditions and as a resistive-type SFCL, IGBT-Bridgetype SFCL during both symmetrical and asymmetrical fault is discussed. The deviation of Voltage has considerably reduced in both case 1 and case 2 as compared to case 1 in which SFCL has not been used. This is due to reduction in fault current at the bus terminal. The fig 4.8 shows the complete DFIG based model using SFCL.  The Universal Bridge block implements a universal three-phase power converter that consists of up to six power switches connected in a bridge configuration. First one is acting as rectifier and second as inverter model. It uses forced commutated devices like IGBT forming three bridge arms. The Snubber Resistance is kept to be 1000 ohms. The snubber capacitance, in farads (F) is set to inf to get a resistive snubber with internal resistance of the selected device 1 x 10 -3 ohms. Pulses to the grid side converter and rotor side converter is provided individually by a controlling system as shown in figure 4. Discrete 3-phase PWM Generator are used to generate pulses. Sampling time (Ts) is taken to be 5.000 x 10 -3 .

V. RESULTS
The comparison of the DFIG model without using any controller, with DC resistive controller and DFIG -Bridge Type SFCL has been shown below: The model is being simulated for 0.5, 1, 1.5, 2 seconds. The corresponding values of power deviation, terminal Voltage deviation, current deviation and speed deviation has been recorded and plotted against time axis.

VI. CONCLUSION
This study proposes the use of IGBT-bridge-Type SFCL to improve the transient stability of a DFIG based wind power system. From the simulation results, the following assessments can be noted: (i) The proposed SFCL design with gated IGBT (IGBTbridge-Type SFCL) is able to enhance the transient stability of the DFIG system for both the symmetrical and asymmetrical faults more efficiently than the DC Resistive type SFCL.
(ii) The values of Voltage deviation at the bus terminal is 8.223 e -8 % for IGBT-bridge-Type SFCL and it is 14.4 e -8 % for Resistive type SFCL when the DFIG model is simulated for 2 seconds. (iii) The huge Voltage sag has been considerably reduced by reduction of high level of current to 0.0004401 % in IGBT-bridge-Type SFCL from 0.0004624% in resistive type SFCL. (iv) Huge Voltage sag and high levels of fault current are significantly suppressed by the IGBT-bridge-Type SFCL. Thus proposed SFCL has caused significant improvement in transient stability keeping the deviation in active and reactive power during faults to minimum level. The use of fast switching, forced commutated device like IGBT shunted with a variable resistance, can improve the performance of the system significantly during abnormal conditions. Further the enhancement of this design can lead to a better output and can make this system to work for low Voltage ride through LVRT also.