Three-phase AC-DC Converter for Direct-drive PMSG-based Wind Energy Conversion System

Kumar Abhishek Singh; Ayushi Chaudhary; Kalpana Chaudhary

网刊加载中。。。

使用Chrome浏览器效果最佳，继续浏览，你可能不会看到最佳的展示效果，

确定继续浏览么?

复制成功，请在其他浏览器进行阅读

Abstract

In this paper, a wind energy conversion system (WECS) is presented for the electrification of rural areas with wind energy availability. A three-phase AC-DC converter based on a bridgeless Cuk converter is used for power extraction from the permanent magnet synchronous generator (PMSG). The bridgeless topology enables the elimination of the front-end diode bridge rectifier (DBR). Moreover, the converter has fewer components, simple control, and high efficiency, making it suitable for a small-scale WECS. A squirrel cage induction motor (SCIM) is used to emulate a MOD-2 wind turbine to implement the PMSG-based WECS. A direct-drive eight-pole PMSG is used in this study; thus, a low-input-voltage system is designed. The converter is designed to operate in the discontinuous inductor current mode (DICM) for inherent power factor correction (PFC) and the maximum power point tracking (MPPT) is achieved through the tip-speed ratio (TSR) following. The performance of the developed system is analyzed through simulation, and a 500 W hardware prototype is developed and tested in different wind speed conditions.

Keywords

Permanent magnet synchronous generator （PMSG); three-phase AC-DC converter; wind energy conversion system (WECS); wind turbine emulator

I. Introduction

EXTENSIVE studies have been conducted on the adverse effects of renewable energy integration into the power grid [

1]-[5]. Although these shortcomings can be overcome using advanced control strategies [6]-[8] and converter modification [9]-[11], a more viable solution to this problem is to integrate renewable sources into DC microgrids [2], [12]. In addition, the majority of household loads are DC loads in nature, and the end-to-end efficiency with a DC microgrid is higher than that with an AC microgrid with DC loads [13]. An economically suitable and efficient DC converter is required for the maximum power point tracking (MPPT) and energy flow control to integrate the renewable sources into a DC microgrid. In this paper, an efficient three-phase AC-DC converter is presented for the power control of wind turbine emulator driven permanent magnet synchronous generator (PMSG). The details of the converter operation with a self-excited induction generator (SEIG) are provided in [14].

The grid-connected wind energy conversion system (WECS) is based on partial-scale power converters in the case of a doubly fed induction generator (DFIG) and full-scale power converters in the cases of PMSG and SEIG [

1]. This is the main reason for the popularity of DFIG for WECS. However, with increasing power levels of wind turbines, the partial-scale power converter rating is redundant because many power converters in parallel are used for grid integration. Moreover, the full-scale power converters are more versatile, as they can be used for active stall techniques under high wind speed conditions [15]. The higher efficiency and broader range of operations have made PMSG more popular in the past decade. Moreover, the low speed of wind turbines leads to bulkier DFIGs with a large number of poles or the use of a gearbox. The gearbox, brush contact, and rotor losses can be eliminated in a PMSG-based WECS, thus increasing the system reliability and lowering the overall cost [6], [8]. The PMSG is considered more relevant for rural applications in this study because of its robustness and low maintenance cost.

Power electronic converters are the most vulnerable part of the WECS; thus, a robust, efficient, and inexpensive AC-DC converter is required [

1], [16]. In most conventional PMSG-based WECSs, a front-end diode bridge rectifier (DBR) is used for rectification, followed by a DC-DC converter for power control and MPPT [6], [7], [17]. The decrease in efficiency owing to the front-end DBR and the popularity of bridgeless topology have recently motivated researchers to work toward a more efficient converter design for WECS. Reference [9] proposes a bridgeless boost rectifier for WECS, which requires connections from all six terminals of the PMSG and six switches for rectification. Reference [18] proposes a compact AC-DC converter, which has a larger number of switches and uses an isolation transformer, thus increasing the complexity of control and cost. Reference [10] presents a three-phase AC-DC converter using space vector modulation, but it requires current sensors for power factor correction (PFC). All these studies require sensing of the generator voltage and current to control the power. A single-stage DC to three-phase AC converter is presented in [19], but it cannot be used for WECS because it is unidirectional in nature and thus cannot perform AC-DC conversion. A bi-directional three-phase AC-DC converter is presented in [20], which employs a boost strategy at both the AC and DC sides and has three stages of conversion, resulting in increased component count and complexity. Reference [21] proposes a unique three-phase AC-DC conversion strategy using coupled inductors, but it requires two parallel converters and has high weight and cost owing to the large number of inductors used. Table I presents a comparison of the proposed converter with other typical converters used in three-phase AC-DC conversion.

TABLE I Comparison of Proposed Converter with Other Typical Converters

Converter	No. of switches	No. of diodes	No. of inductors	No. of capacitors	Complexity	Power flow
SVPWM controlled TPTPR [22]	12	0	3	2	Complex	Bi-directional
Bi-directional three-phase AC-DC converter [23]	10	0	7	2	Complex	Bi-directional
Three-phase unfolded circuit based converter [24]	12	6	3	1	Very complex	Bi-directional
Matrix based AC-DC converter [25]	8	2	6	4	Complex	Unidirectional
Multiport converter with independent control [26]	12	0	9	2	Complex	Bi-directional
Proposed converter	3	3	3	2	Simple	Unidirectional

Note: SVPWM stands for space vector pulse width modulation; and TPTPR stands for three-port three-phase rectifier.

This study aims to implement this converter in the WECS for rural and isolated areas with a reduced component count and high efficiency. The topology of the proposed three-phase AC-DC bridgeless Cuk converter is illustrated in Fig. 1.

Fig. 1 Topology of proposed three-phase AC-DC bridgeless Cuk converter.

The main contributions of this study are summarized as follows.

1) The developed converter has less switches and a simple control, and can operate in a broader voltage range.

2) A methodology for developing a wind turbine emulator using a squirrel-cage induction motor (SCIM) with field-oriented control (FOC) is presented.

3) The tip-speed ratio (TSR) following the MPPT technique is used.

The remainder of this paper is organized as follows. The design and operation of converter are briefly discussed in Section II. Section III describes the WECS setup used in this study. Simulation results with brief discussions are presented in Section IV. Section V presents the hardware results. The conclusions are presented in Section VI.

II. Bridgeless Cuk Converter

A. Circuit Description

As shown in Fig.1, the proposed converter can operate with a three-phase generator with the input inductors $L_{1} - L_{3}$ in series with each phase, followed by the typical Cuk converter topology for each phase. The switches $Q_{1} - Q_{3}$ are used to energize the input inductors. The capacitors $C_{1} - C_{3}$ are then charged through the input inductors through the diodes $D_{1} - D_{3}$ , respectively. The capacitors discharge their energy to the output capacitor $C_{o}$ through the output inductors $L_{11}$ , $L_{22}$ , and $L_{33}$ .

The proposed converter has the following advantages over the conventional bridgeless Cuk converter.

1) There are no diodes connected directly to the source; thus, generator leakage inductance can be used effectively as the input inductance.

2) The switching control is simple and thus needs less components, as a single gate driver can be used to switch all three switches.

3) The topology of three-phase bridgeless Cuk converter ensures only 50% working cycle for capacitors, output inductors, and diodes.

4) The back diodes of the switches are used in the negative half-cycle of the source voltage in the respective phase, which leads to further component reduction.

B. Converter Operation

The converter is designed to operate in the discontinuous inductor current mode (DICM) of the input inductances. This ensures an inherent PFC and allows the converter to operate more efficiently by drawing only active power from the generator. The converter operation is defined in the following three modes.

1)　Mode-1

In Mode-1 of the converter operation depicted in Fig. 2, $Q_{1} - Q_{3}$ are switched “ON” simultaneously, which allows the input inductors to be charged in a positive or negative direction depending on the voltage of the respective phase. In Fig. 2, phases R and Y have positive voltages, and phase B has a negative voltage. As the output inductors are in the continuous inductor current mode (CICM), the energy of output inductors $L_{11}$ and $L_{22}$ is discharged to capacitor $C_{o}$ . The output inductor $L_{33}$ is fully discharged as phase B has a negative voltage; thus, the inductor is not charged.

Fig. 2 Mode-1 of converter operation.

2)　Mode-2

Figure 3 depicts Mode-2 of the converter operation. In this mode, all the switches $Q_{1} - Q_{3}$ are turned “OFF”, and the energy of input inductors L₁-L₃ is discharged to capacitors $C_{1} - C_{3}$ through diodes $D_{1} - D_{3}$ . In this mode, the energy of input inductors in phases R and Y with positive voltages is discharged to capacitors $C_{1}$ and $C_{2}$ . The energy of input inductor in phase B with negative voltage is also discharged to the same capacitors through the back diode of the switch $Q_{3}$ . In this mode, the energy of output inductors $L_{11}$ and $L_{22}$ is discharged to the output capacitor through diodes $D_{1}$ and $D_{2}$ . This mode ends when the input inductors are fully discharged.

Fig. 3 Mode-2 of converter operation.

3)　Mode-3

The mode-3 of converter operation is depicted in Fig. 4. In this mode, the switches $Q_{1} - Q_{3}$ remain in the “OFF” state. All the input inductors are fully discharged. The energy of output inductors is continuously discharged to the output capacitor through the diodes $D_{1}$ and $D_{2}$ because these inductors operate in the CICM.

Fig. 4 Mode-3 of converter operation.

C. Converter Analysis

The converter is designed to keep the input inductor in the deep DICM. Thus, a converter analysis is also needed with the same consideration. To find the converter voltage gain $V_{o} / V_{i n}$ , it is necessary to find the voltages across each input inductor.

V_{R} = V_{m} s i n (ω t)

(1)

V_{Y} = V_{m} s i n (ω t + 2 π / 3)

(2)

V_{B} = V_{m} s i n (ω t - 2 π / 3)

(3)

where $V_{R}$ , $V_{Y}$ , and $V_{B}$ are the voltages of phases R, Y, and B, respectively; $V_{m}$ is the amplitude of the three phases; and $ω$ is the speed angle frequency.

By adding the squares of $V_{R}$ , $V_{Y}$ , and $V_{B}$ , we can obtain:

V_{R}^{2} + V_{Y}^{2} + V_{B}^{2} = \frac{3}{2} V_{m}^{2}

(4)

Now, as the energy absorption is performed in Mode-1 of the converter operation, the energy absorbed in Mode-1 by input inductors in “dt”, i.e., $d E$ , can be written as:

d E

= V_{L} I_{L} d t

= V_{L} \frac{V_{L} t}{L} d t

(5)

where $V_{L}$ is the three-phase average root mean square (RMS) voltage; and $I_{L}$ is the current at time t.

Thus, the total energy $Δ E$ absorbed in each phase in time DT can be expressed as:

Δ E = \int_{0}^{D T} \frac{V_{L}^{2} t}{L} d t

(6)

Δ E = \frac{V_{L}^{2} D^{2} T^{2}}{2 L}

(7)

where D is the duty cycle; and T is the total time cycle.

The total energy absorbed can be calculated using (4) and (7). The energy of inductors with negative currents are also discharged to capacitors. This energy is then transferred to the output capacitor. Thus, the total energy can be written as:

Δ E_{T} = \frac{D^{2} T^{2}}{2 L} (V_{R}^{2} + V_{Y}^{2} + V_{B}^{2})

(8)

Δ E_{T} = \frac{3 D^{2} T^{2}}{4 L} V_{m}^{2}

(9)

Equation (9) represents the total input energy to the converter in time DT. Notably, the energy absorption occurs only for time DT, as the total time cycle is T. The input and output power $P_{i n}$ and $P_{o}$ can be written as:

P_{i n} = \frac{Δ E_{T}}{T}

(10)

P_{i n} = \frac{3 D^{2} T V_{m}^{2}}{4 L}

(11)

P_{o} = η P_{i n}

(12)

P_{o} = \frac{V_{o}^{2}}{R_{L}}

(13)

where $η$ is the coefficient of power. Thus, the output voltage $V_{o}$ can be expressed as:

V_{o} =

\sqrt[]{P_{o} R_{L}}

\sqrt[]{η P_{i n} R_{L}}

(14)

V_{o} =

\sqrt[]{\frac{3 η T}{L}} \frac{D V_{m}}{2}

(15)

D. Small-signal Analysis

The converter is analyzed using the current-injected equivalent circuit method [

27]. A single-phase model of the converter is used for the analysis and to obtain the small-signal transfer function. In Fig. 5, the circuit is divided into two parts for analysis. The part of the circuit from the input to

x x'

is the input, whereas the part of the circuit from

y y'

to the output is the output. For the first part of the circuit,

V_{c}

is the output voltage and

V_{i n}

is the input voltage. The average inductor current injected to capacitor C, i.e.,

i_{a v g}

, is given by:

i_{a v g} = D_{2} I_{1}

(16)

Fig. 5 Equivalent circuit for small-signal analysis.

where $D_{2} = 1 - D_{1}$ , and D₁ is the duty cycle of Q₁; and $I_{1}$ is the input inductor current.

The derivative of the input inductor current is given by:

\frac{L_{1} d I_{1}}{d t} = V_{i n} - D_{2} V_{c}

(17)

Let the effective load on the capacitor C be $R_{e}$ . Then, we can obtain:

V_{c} = i_{a v g} (C ∥ R_{e})

(18)

R_{e} = V_{o} / I_{o}

(19)

where $I_{o}$ is the output current.

From the output part of the circuit, the derivative of the inductor current is given by:

\frac{L_{o} d I_{o}}{d t} = D_{1} V_{c} - V_{o}

(20)

Perturbing (16)-(19) around the steady state and taking the Laplace transform, the small-signal linear equations are determined as:

{\hat{i}}_{a v g} = D_{2} {\hat{i}}_{1} - I_{1} {\hat{d}}_{1}

(21)

s L_{1} {\hat{i}}_{1} = {\hat{V}}_{i n} - D_{2} {\hat{V}}_{c} + V_{c} {\hat{d}}_{1}

(22)

{\hat{v}}_{c} = R {\hat{i}}_{a v g} - \frac{I_{1} D_{2} R {\hat{i}}_{o}}{I_{o}}

(23)

where $R = V_{o} / (s C V_{o} + I_{o})$ ; and ${\hat{d}}_{1}$ is the perturbed value of $D_{1}$ .

Solving (5)-(7) yields:

{\hat{V}}_{c} (\frac{1}{R} + \frac{D_{2}^{2}}{s L_{1}}) = \frac{D_{2}}{s L_{1}} {\hat{v}}_{i n} + \frac{V_{o} D_{2}}{s L_{1} D_{1}} {\hat{d}}_{1} - \frac{1}{D_{1}} {\hat{i}}_{o}

(24)

In the output part of the converter, the state of the output capacitor is not perturbed, as the converter is employed in a DC microgrid of 230 V. The stiff DC voltage at the output is constant. Thus, (20) is perturbed to obtain:

s L_{2} {\hat{i}}_{o} = D_{1} {\hat{v}}_{c} + V_{c} {\hat{d}}_{1}

(25)

Thus, solving (24) and (25) yields:

(s L_{2} + \frac{1}{X}) {\hat{i}}_{o} = \frac{D_{1} D_{2}}{s X L_{1}} {\hat{v}}_{i n} + (\frac{V_{o} D_{2}}{s X L_{1}} + \frac{V_{o}}{D_{1}}) {\hat{d}}_{1}

(26)

where $X = 1 / R + D_{2}^{2} / (s L_{1})$ .

As the output power is proportional to the output current $I_{o}$ , the control-to-output current transfer function is presented here by substituting ${\hat{v}}_{i n} = 0$ in (26).

\frac{{\hat{i}}_{o}}{{\hat{d}}_{1}} = \frac{s^{2} L_{1} V_{o}^{2} C D_{1}^{- 1} + s L_{1} V_{o} I_{o} D_{1}^{- 1} + V_{o}^{2} D_{2} D_{1}^{- 1}}{s^{3} L_{1} L_{2} V_{o} C + s^{2} L_{1} I_{o} L_{2} + s L_{2} D_{2}^{2} V_{o} + 1}

(27)

E. Converter Design

The converter is designed to maintain the input inductor in the DICM and keep the input inductance of the three-phase circuit maximum at the power line frequency. The design parameters assumed in this study are listed in Table II. As the converter is intended to be placed near the wind turbine, the noise produced at the audible frequency is not considered as a design factor. The operating frequency f of 10 kHz is selected. The advantage of the direct coupling of PMSG to the turbine is the elimination of the gearbox and the consequent decrease in cost, but the commercially available PMSGs have a low constant voltage. An eight-pole PMSG is used in this study; thus, for better power extraction from the generator, the converter is designed to operate at a low input RMS voltage of 100 V.

TABLE II Design Parameters

Parameter	Value
$V_{i n}$	100 V
$P_{o}$	500 W
$V_{o}$	350 V
f	10 kHz

It is convenient to perform the calculations in per-phase quantities. The design methodology is as follows: ① the maximum input voltage per phase is $V_{p p} = \sqrt[]{2} V_{m} / \sqrt[]{3} = 81.4 V$ ; ② the maximum input current $I_{m a x}$ = $V_{p p} D T_{s} / L_{1}$ ; ③ the saturation current $I_{m a x}$ of the input inductors is taken as 8 A; ④ the maximum operating duty cycle $D_{m a x}$ for the DICM operation is taken as 70%; and ⑤ the load R_L for the converter is taken as 250 $Ω$ .

The input inductance is calculated as:

L_{1} = \frac{V_{p p} D_{m a x} T_{s}}{I_{m a x}} = 712 μ H

(28)

The input inductance is taken to be 560 $μ H$ (commercially available value) to maintain the inductor in the deep DICM.

The critical value of $K_{a}$ is calculated as:

K_{a, c r i t i c a l} = \frac{1}{2 {(1 + M)}^{2}} = \frac{1}{2 {(1 + \frac{V_{o}}{V_{i n}})}^{2}}

(29)

The chosen values of the output and input voltages are 350 V and 100 $\sqrt[]{2}$ V, respectively. From (2), $K_{a, c r i t i c a l}$ is calculated to be 0.0412; thus, for the DICM mode, a lower value of $K_{a}$ is taken as 0.03. The value of the equivalent inductance $L_{e q}$ is calculated as:

\frac{1}{L_{e q}} = \frac{1}{L_{1}} + \frac{1}{L_{11}}

(30)

L_{e q} = \frac{R_{L} K_{a} T_{s}}{2} = 375 μ H

(31)

Taking the value of $L_{1}$ from (28) and using (30) and (31), the output inductance $L_{11}$ can be calculated as:

L_{11} = \frac{L_{1} L_{e q}}{L_{1} - L_{e q}} = 1135 μ H \approx 1120 μ H

(32)

The capacitance $C_{1}$ is taken as 1 $μ F$ , where the resonant frequency of the LC oscillation with the input inductor, output inductor, and combined $L_{1} + L_{11}$ is not near the switching frequency or the input power frequency, which ranges from 25 Hz to 100 Hz.

The DC link capacitance is selected to obtain the minimum voltage ripple at the rated output power, and the output capacitance is chosen as a reasonable value of 1000 $μ F$ . At this value, the voltage ripple is found to be less than 2%. Table III summarizes the converter parameters selected for this study.

TABLE III Converter Parameters

Converter parameter	Value
$L_{1}$	560 $μ H$
$L_{11}$	1120 $μ H$
$C_{1}$	1 $μ F$
$C_{o}$	1000 $μ F$

III. WECS Architecture

To test the proposed converter, a WECS is realized, which consists of a wind turbine emulator, the proposed converter, and a DC load. The emulator is developed using an SCIM driven by decoupled FOC. In this study, a MOD-2 turbine type is emulated, as it has a simpler mathematical model and a higher value of TSR, denoted as $λ$ . A higher TSR is more suitable for the direct-drive PMSG-based WECS because a higher rotor speed is achieved at all wind speeds.

A. MOD-2 Turbine Model

The power coefficient $C_{p}$ , which depends on the TSR $λ$ and blade pitch $β$ , determines the amount of energy extracted from the wind by the turbine, denoted as $P_{t u r b i n e}$ . For a blade of radius R rotating with an angular velocity of $ω_{r}$ at wind speed $v$ , the value of TSR $λ$ is given as:

λ = \frac{ω_{r} R}{v}

(33)

P_{t u r b i n e} = \frac{1}{2} C_{p} ρ A v^{3}

(34)

Based on the calculations performed by Betz, the maximum amount of wind power that can be extracted using a turbine is expressed as:

P_{m a x} = \frac{16}{27} \frac{1}{2} ρ A v^{3}

(35)

By gathering the field data and plotting $C_{p}$ - $λ$ curves, according to [

28], a nonlinear model can be derived as:

C_{p} = C_{1} (C_{2} - C_{3} β - C_{4} β^{x} - C_{5}) e^{- C_{6} (λ, β)}

(36)

where for MOD-2 wind turbine, the values of $C_{1}$ - $C_{6}$ are selected as $C_{1} = 0.5$ , $C_{2} = 116 / λ_{i},$ $C_{3} = 0.4$ , $C_{4} = 0$ , $C_{5} = 5$ , and $C_{6} = 21 / λ_{i}$ , and $λ_{i} = 1 / (λ + 0.08 β) - 0.035 / (β^{3} + 1)$ .

B. SCIM Control

A 5 kW, two-pole SCIM is selected to emulate the wind turbine. The control actions are implemented using CP1104 dSPACE. Using the turbine model given in Section III-A, a MATLAB model is developed to generate the torque reference for the SCIM. Using (36), the torque reference $T_{r e f}$ can be calculated at rotor speed $ω_{g e n}$ as:

T_{r e f} = \frac{1}{2} \frac{C_{p} ρ A v^{3}}{ω_{g e n}}

(37)

The stator and rotor leakage inductances $L_{s}$ and $L_{r}$ , mutual inductance $L_{m}$ , steady-state rotor flux $ψ_{r}$ , and stator and rotor resistances $R_{s}$ and $R_{r}$ are calculated by performing no-load and block-rotor tests on the SCIM. The FOC is implemented on the SCIM as follows.

1) The stator currents $I_{a}, I_{b},$ and $I_{c}$ are measured and converted to the d-q reference frame as $I_{d}$ and $I_{q}$ using the Park transformation.

2) $I_{q}$ is used to calculate the slip speed $ω_{s}$ in the SCIM using (38) and (39).

ω_{s} = \frac{L_{m} I_{q}}{τ ψ_{r}}

(38)

τ = \frac{L_{m} + L_{r}}{R_{r}}

(39)

where $τ$ is the motor time constant; and $R_{r}$ is the rotor resistance.

This calculated slip speed is added to the measured rotor speed of the machine to obtain the electrical angle of the stator currents $ω_{e}$ :

ω_{e} = ω_{r} + ω_{s}

(40)

By using the measured value of $I_{d}$ , the value of rotor flux $ψ_{r}$ can be estimated by:

ψ_{r} = \frac{L_{m}}{1 + s τ} I_{d}

(41)

For the implementation using a digital system, the Laplace transform is changed into the z-domain using the values given in Table IV.

ψ_{r} = \frac{0.0003902 z}{z - 0.99967}

(42)

TABLE IV Motor Parameters

Motor parameter	Values
$L_{s}$	0.0138 H
$L_{r}$	0.0138 H
$R_{S}$	1.09 Ω
$R_{r}$	3.917 Ω
$L_{m}$	1.182 H
$τ$	0.305285 s

The torque of the machine needs to be estimated accurately to implement the torque control for turbine emulation.

T_{e} = \frac{3}{2} P \frac{L_{m}}{L_{m} + L_{r}} (ψ_{r} I_{q})

(43)

where P is the number of pole pairs in the SCIM and $P = 1$ .

The values of different motor parameters used in (38), (39), and (43) are listed in Table IV. The values of $I_{d}$ and $I_{q}$ are DC currents in nature; thus, PI controllers are deemed sufficient to control the stator currents by generating a voltage reference for the VSI. The torque reference is generated using (37), and the flux reference is set as a value calculated by performing a no-load test on the SCIM. In this study, the flux reference is set as 1.01 Wb. A proportional-integral (PI) controller with a proportional gain $K_{p}$ of 1.2 and an integral gain $K_{i}$ of 50 is used for both the torque and flux controllers.

C. MPPT and System Description

The maximum power extracted from an ideal wind turbine is given by (35). Ideally, a maximum of 59.26% of the input wind power can be extracted. This makes the use of an MPPT system pivotal in a WECS. For fixed-blade-pitch turbines used in small-scale wind generation, the TSR value determines the value of $C_{p}$ ; thus, the TSR-MPPT technique is used in this study.

The schematic of the wind energy emulation and coversion system is shown in Fig. 6. The system description for motor control is shown in Fig. 7. The MPPT technique is simple to implement using a PI controller with saturation at 0-0.7. The PI controller has back calculation as an anti-windup method. The values of $K_{p}$ and $K_{i}$ used for the TSR-MPPT control are 0.2 and 5, respectively. For the MOD-2 turbine emulated in this study, the optimum TSR value at the $0 °$ blade pitch is observed to be 8.035.

Fig. 6 Schematic of wind energy emulation and conversion system.

Fig. 7 System description for motor control.

IV. Simulation Results

The system represented in Fig. 6 is simulated in MATLAB/Simulink version 8.9, with a variable step and the ode23tb solver. The simulated system performance is similar to that obtained through the experimental verification of the system. The calculations for flux estimation, torque reference generation, and the PI controller are performed at a fixed step time of 0.0001 s. This is the same as the step time of dSPACE, which is used in the hardware setup.

The simulated results of the three-phase input voltages and currents of the steady-state converter are shown in Fig. 8. The simulated result shows the inherent PFC at the converter input owing to the DICM operation.

Fig. 8 Three-phase input voltages and currents of steady-state converter.

The starting characteristics and response to a step change in wind speed are analyzed below. A simulation for a step change in wind speed from 10 m/s to 15 m/s at $t = 0.5 s$ is performed, and the results of TSR following and controller output (duty cycle) for MPPT are presented in Fig. 9. A sudden decrease in the controller output is observed at $t = 0.5 s$ owing to a steep decrease in the value of TSR $λ$ , as the generator speed is constant during this time period. This decrease in the value of the TSR prompts the PI controller to decrease its output and then increase the generator speed to match the reference TSR value.

Fig. 9 TSR following and controller output for MPPT.

The rotor speed and electromagnetic torque of the SCIM are shown in Fig. 10. A sudden rise in the speed and torque could be observed at $t = 0.5 s$ in response to the increased wind speed. During the starting of the machine, a rise in the torque reference is observed as a result of the increase in $C_{p}$ at the optimal speed of the generator for the given wind speed. This increase in $C_{p}$ leads to a high torque reference during the starting of the machine. The simulation for the VSI control is performed for different DC link voltages, and it is observed to work satisfactorily even at lower DC link voltages when the operating speed of the emulator is less than 1500 r/min. Instead of a 590 V DC link for full-range SCIM operation, a 400 V DC link is used in the simulation and in the hardware setup.

Fig. 10 Rotor speed and electromagnetic torque of SCIM.

The DC output voltage of the converter and the voltage generation of PMSG at a step change in wind speed from 10 m/s to 15 m/s at $t = 0.5 s$ is shown in Fig. 11. A large output capacitor regulates the DC voltage ripple below 50% of the mean output DC voltage level.

Fig. 11 DC output voltage of converter and voltage generation of PMSG.

The efficiency of the three-phase AC-DC conversion system used in [

8] for the WECS is compared with that of the proposed converter. The efficiency curves with changes in the converter duty cycle is shown in Fig. 12, where the efficiency of the uncontrolled DBR is also shown. The efficiency of the proposed converter attains values similar to the DBR with the duty cycle of approximately 35%.

Fig. 12 Comparative efficiency of different converters.

V. Hardware Results

The developed converter is tested using a PMSG (BLS-143) driven by a wind turbine emulator, which is developed using a three-phase 5.5 kW induction motor (1722 J). The components used to develop the converter are presented in Table V. The input inductance and output inductance are selected as 560 $μ H$ and 1120 $μ H$ , respectively, as per the design described in Section II-V. Thus, nine inductors of 560 $μ H$ are used in this study, with three inductors per phase. The emulator and converter are controlled using a CP1104 d-SPACE.

TABLE V Component Description

Component	Part number	Quantity
Inductors	PCV-2-564-08L	9
Capacitor (1 $μ F$ )	MKP C.4C	3
Capacitor (1 mF)	ALC70C 102 FP	1
Insulated gate bipolar transistors (IGBTs)	K40H1203	3
Diodes	DSEI60-10A	3

The steady-state operation of the converter at the wind speed of 12 m/s is shown in Fig. 13. A resistive load of 100 $Ω$ is used for the hardware implementation and simulation of the system. The output power of converter is measured to be 316 W with an input power of 330 W; thus, the efficiency is more than 95%. An eight-pole PMSG is used in this study to obtain a higher voltage at a low rotor speed of 972 r/min.

Fig. 13 Steady-state operation of converter at wind speed of 12 m/s.

During the wind speed variation, the load resistance is maintained constant with the controller tracking the optimum TSR value. This working condition helps maintain the system operation at the maximum power point (MPP) even after the induced step change in the wind speed.

Figure 14 shows the system performance at a sudden drop in wind speed from 12 m/s to 10 m/s. It is evident from the hardware results that the WECS attains the MPPT with an increase and decrease in wind speed. A smooth transition in the power level occurs in both the emulator and converter.

Fig. 14 System performance at a sudden drop in wind speed from 12 m/s to 10 m/s.

A step change in wind speed from 10 m/s to 12 m/s is simulated, and the result is presented in Fig. 15. A smooth transition in the power level is observed for small step changes in the wind speed. An increase in the input current of converter can be observed, which results in an increase in the output DC voltage.

Fig. 15 System performance at a step change in wind speed from 10 m/s to 12 m/s.

To test the system in the wind shear condition, a step change in wind speed from 10 m/s to 15 m/s is introduced in the system. The results are shown in Fig. 16. At the instant of wind speed change, a sudden drop in the converter duty is observed. As the rotor speed is constant owing to inertia and wind speed increases, the calculated value of $λ$ decreases sharply. This decrease in the value of $λ$ leads to a drop in the converter duty; however, the system attains the MPPT condition and reaches a steady state within 0.3 s.

Fig. 16 System performance in wind shear condition at a change in wind speed from 10 m/s to 15 m/s.

The hardware results for the DICM operation of the converter with a duty cycle lower than 70%, at a high wind speed of 15 m/s, are presented in Fig. 17. The total input power is measured using the mathematical function in the digital storage oscilloscope (DSO), which is calculated as three times the mean value. All the measurements performed in this study have a load resistance of 100 $Ω$ at the output of the converter. The efficiency of the converter is measured as 95.184%. The conversion efficiency of the designed converter is observed to be approximately 95% for all the operating conditions in the DICM, above an output power of 200 W.

Fig. 17 DICM operation of converter with duty cycle lower than 70%.

The hardware results for CICM operation of the converter with a duty cycle higher than 70% are shown in Fig. 18. The input current of converter is nonzero at every nonzero voltage. The peak current value saturates the inductors; thus, slight noise is observed in the input currents. The converter input power is 639 W, and the output power is 576 W, and thus, the converter efficiency is decreased to 90.14%.

Fig. 18 CICM operation of converter with duty cycle higher than 70%.

Figure A1 in Appendix A shows the laboratory setup for the wind turbine emulation and the PMSG-based WECS. The turbine is emulated using a three-phase SCIM that is mechanically coupled with the BLS-143 PMSG. The developed converter is fed with a variable-frequency three-phase supply from the PMSG, which is converted to a DC supply feeding the resistive load.

VI. Conclusion

A PMSG-based WECS using a new three-phase AC-DC converter is designed and tested in this study. The developed WECS is designed and tested as a standalone system. It is envisaged that the system can be easily integrated with a photovoltaic system or battery bank. The functioning of the emulator is tested by employing the TSR-MPPT technique, and the system is experimentally validated at different wind speeds. The elimination of the DBR and the compatibility of the designed converter with the PMSG make the system more efficient with fewer components. The simulation and laboratory prototype confirm the effectiveness of using this system in WECSs.

Appendix

Appendix A

Fig. A1 Laboratory setup for wind turbine emulation and PMSG-based WECS.

References

G. Chen and X. Cai, “Reconfigurable control for fault-tolerant of parallel converters in PMSG wind energy conversion system,” IEEE Transactions on Sustainable Energy, vol. 10, no. 2, pp. 604-614, Apr. 2019. [Baidu Scholar]

A. W. Bizuayehu, D. Z. Fitiwi, and J. P. Catalao, “Advantages of optimal storage location and size on the economic dispatch in distribution systems,” IEEE PES General Meeting, vol. 2016, no. 3, pp. 1336-1345, Jul. 2016. [Baidu Scholar]

Y. Tan, H. Zhang, and Y. Zhou, “Fault detection method for permanent magnet synchronous generator wind energy converters using correlation features among three-phase currents,” Journal of Modern Power Systems and Clean Energy, vol. 8, no. 1, pp. 168-178, Jan. 2020. [Baidu Scholar]

A. A. A. Radwan and Y. A. R. I. Mohamed, “Grid-connected wind-solar cogeneration using back-to-back voltage-source converters,” IEEE Transactions on Sustainable Energy, vol. 11, no. 1, pp. 315-325, Jan. 2020. [Baidu Scholar]

V. Nayanar, N. Kumaresan, and N. A. Gounden, “A single-sensor-based MPPT controller for wind-driven induction generators supplying DC microgrid,” IEEE Transactions on Power Electronics, vol. 31, no. 2, pp. 1161-1172, Feb. 2016. [Baidu Scholar]

J. Chen, T. Lin, C. Wen et al., “Design of a unified power controller for variable-speed fixed-pitch wind energy conversion system,” IEEE Transactions on Industrial Electronics, vol. 63, no. 8, pp. 4899-4908, Aug. 2016. [Baidu Scholar]

R. Errouissi, A. Al-Durra, and M. Debouza, “A novel design of PI current controller for PMSG-based wind turbine considering transient performance specifications and control saturation,” IEEE Transactions on Industrial Electronics, vol. 65, no. 11, pp. 8624-8634, Nov. 2018. [Baidu Scholar]

B. Rupam and G. Bhuvaneswari, “Low-voltage ride-through of a synchronous wind energy conversion system,” IEEE Transactions on Industry Applications, vol. 56, no. 1, pp. 752-762, Feb. 2020. [Baidu Scholar]

J. S. Guimaraes, B. R. D. Almeida, F. L. Tofoli et al., “Three-phase grid-connected WECS with mechanical power control,” IEEE Transactions on Sustainable Energy, vol. 9, no. 4, pp 1508-1517, Oct. 2018. [Baidu Scholar]

X. Guo, Y. Yang, B. Wang et al., “Generalized space vector modulation for current source converter in continuous and discontinuous current modes,” IEEE Transactions on Industrial Electronics, vol. 67, no. 11, pp. 9348-9357, Nov. 2020. [Baidu Scholar]

R. Barrera-Cardenas and M. Molinas, “Comparative study of wind turbine power converters based on medium-frequency AC-link for offshore DC-grids,” IEEE Journal of Emerging and Selected Topics in Power Electronics, vol. 3, no. 2, pp. 525-541, Jun. 2015. [Baidu Scholar]

M. Abbasi and J. Lam, “A step-up transformerless, ZV-ZCS high-gain DC/DC converter with output voltage regulation using modular step-up resonant cells for DC grid in wind systems,” IEEE Journal of Emerging and Selected Topics in Power Electronics, vol. 5, no. 3, pp. 1102-1121, Sept. 2017. [Baidu Scholar]

M. Nasir, H. A. Khan, A. Hussain et al., “Solar PV-based scalable DC microgrid for rural electrification in developing regions,” IEEE Transactions on Sustainable Energy, vol. 9, no. 1, pp. 390-399, Jan. 2018. [Baidu Scholar]

A. K. Singh, A. K. Mishra, K. K. Gupta et al., “An integrated converter with reduced components for electric vehicles utilizing solar and grid power sources,” IEEE Transactions on Transportation Electrification, vol. 6, no. 2, pp. 439-452, Jun. 2020. [Baidu Scholar]

J. M. Guerrero, C. Lumbreras, D. D. Reigosa et al., “Control and emulation of small wind turbines using torque estimators,” IEEE Transactions on Industry Applications, vol. 53, no. 5, pp. 4863-4876, Oct. 2017. [Baidu Scholar]

K. Fischer, T. Stalin, H. Ramberg et al., “Field-experience based root-cause analysis of power-converter failure in wind turbines,” IEEE Transactions on Power Electronics, vol. 30, no. 5, pp. 2481-2492, May 2015. [Baidu Scholar]

R. Nair and G. Narayanan, “Emulation of wind turbine system using vector controlled induction motor drive,” IEEE Transactions on Industry Applications, vol. 56, no. 4, pp. 4124-4133, Aug. 2020. [Baidu Scholar]

J. Khodabakhsh, E. Mohammadi, and G. Moschopoulos, “PMSG-Based wind energy conversion systems integration into DC microgrids with a novel compact converter,” IEEE Access, vol. 8, pp. 83583-83595, May 2020. [Baidu Scholar]

A. Pal and K. Basu, “A Unidirectional single-stage three-phase soft-switched isolated DC-AC Converter,” IEEE Transactions on Power Electronics, vol. 34, no. 2, pp. 1142-1158, Feb. 2019. [Baidu Scholar]

M. Antivachis, “Analysis of a synergetically controlled two-stage three-phase DC/AC buck-boost converter,” CPSS Transactions on Power Electronics and Applications, vol. 5, no. 1, pp. 34-53, Mar. 2020. [Baidu Scholar]

S. Ohn, X. Zhang, R. Burgos et al., “Differential-mode and common-mode coupled inductors for parallel three-phase AC-DC converters,” IEEE Transactions on Power Electronics, vol. 34, no. 3, pp. 2666-2679, May 2019. [Baidu Scholar]

H. Wu, J. Wang, T. Liu et al., “Modified SVPWM-controlled three-port three-phase AC-DC converters with reduced power conversion stages for wide voltage range applications,” IEEE Transactions on Power Electronics, vol. 33, no. 8, pp. 6672-6686, Aug. 2018. [Baidu Scholar]

A. H. Kadam and S. S. Williamson, “A common DC-bus-configured traction motor emulator using a virtually isolated three-phase AC-DC bidirectional converter,” IEEE Access, vol. 9, pp. 80621-80631, Jun. 2021. [Baidu Scholar]

W. W. Chen, R. Zane, and L. Corradini, “Isolated bidirectional grid-tied three-phase AC-DC power conversion using series-resonant converter modules and a three-phase unfolder,” IEEE Transactions on Power Electronics, vol. 32, no. 12, pp. 9001-9012, Dec. 2017. [Baidu Scholar]

A. K. Singh, E. Jeyasankar, P. Das et al., “A single-stage matrix-based isolated three-phase AC-DC converter with novel current commutation,” IEEE Transactions on Transportation Electrification, vol. 3, no. 4, pp. 814-830, Dec. 2017. [Baidu Scholar]

C. Perera, J. Salmon, and G. J. Kish, “Multiport converter with independent control of AC and DC power flows for bipolar DC distribution,” IEEE Transactions on Power Electronics, vol. 36, no. 3, pp. 3473-3485, Mar. 2021. [Baidu Scholar]

P. R. Chetty, “Modelling and analysis of Cuk converter using current injected equivalent circuit approach,” IEEE Transactions on Industrial Electronics, vol. 30, no. 1, pp. 56-59, Feb. 1983. [Baidu Scholar]

Y. H. Han and S. Heier, Grid Integration of Wind Energy Conversion Systems. Chichester: John Wiley & Sons, 2000. [Baidu Scholar]

Address:No.19 Chengxin Avenue, Jiangning District, Nanjing 211106, China

E-mail: mpce@alljournals.cn

Tel:86-25-81093060

Fax:86-25-81093040

Home

Introduction

Editorial Board

For Author

Call For Papers

APC

Sponsor & Publisher

Three-phase AC-DC Converter for Direct-drive PMSG-based Wind Energy Conversion System PDF