Dynamic response and low voltage ride-through enhancement of brushless double-fed induction generator using Salp swarm optimization algorithm.

A brushless double-fed induction generator (BDFIG) has shown tremendous success in wind turbines due to its robust brushless design, smooth operation, and variable speed characteristics. However, the research regarding controlling of machine during low voltage ride through (LVRT) need greater attention as it may cause total disconnection of machine. In addition, the BDFIG based wind turbines must be capable of providing controlled amount of reactive power to the grid as per modern grid code requirements. Also, a suitable dynamic response of machine during both normal and fault conditions needs to be ensured. This paper, as such, attempts to provide reactive power to the grid by analytically calculating the decaying flux and developing a rotor side converter control scheme accordingly. Furthermore, the dynamic response and LVRT capability of the BDFIG is enhanced by using one of the very intelligent optimization algorithms called the Salp Swarm Algorithm (SSA). To prove the efficacy of the proposed control scheme, its performance is compared with that of the particle swan optimization (PSO) based controller in terms of limiting the fault current, regulating active and reactive power, and maintaining the stable operation of the power system under identical operating conditions. The simulation results show that the proposed control scheme significantly improves the dynamic response and LVRT capability of the developed BDFIG based wind energy conversion system; thus proves its essence and efficacy.

1. Introduction

The need for electrical energy is increasing due to the rapidly increasing population and users’ power demand. This continuous increase in power consumption creates several challenges for the electric power companies such as overloading of the existing generating units, transmission lines, transformers, and feeders [1]. Furthermore, the environmental and economic impact of such power systems made them outdated for electricity generation. It is for the mentioned reasons that the trend of relying on fossil fuel-based centralized power systems is declining worldwide and decentralized green energy sources such as wind, solar, and sea tides are considered as potential candidates for power generation. Among the stated green energy options, wind energy has shown quite promising results in terms of reliability and efficiency. However, it has been observed that the increasing penetration of wind energy into the existing power system has introduced few new challenges in terms of grid codes for the advanced generating units, power electronic interfacing devices, and control strategies [2, 3]. In most wind turbines, Double-Fed Induction Generator (DFIG) is utilized to convert the mechanical energy, acquired from the wind, into electrical energy due to its robust design and smooth operation. However, DFIG transfers the generated power to the load by using carbon brushes and slip rings which causes power loss and sparking due to continuous mechanical wear and tear in the mentioned devices [4]. Besides, it provides an inferior low voltage ride through (LVRT) capability as compared to Brushless DFIG which is a comparatively better alternative to the traditional DFIG [5].

Low voltage ride-through, also referred to as fault ride-through, represents the capability of electrical energy generators to remain connected to the power grid during voltage dips [6]. The main motivation behind using generators with good LVRT capability is to prevent widespread loss of generation caused by a short circuit at high voltage (HV) or extremely high voltage (EHV) levels [7]. The winding currents in a generator are responsible to maintain the magnetic field with minimum voltage for its normal operation [8]. However, if this voltage drops below a certain level, the generator may disconnect from the power grid leading to unintentional blackouts [9]. This problem becomes even severe in the case of BDFIG based power system due to the presence of two highly inductive power windings. To avoid the mentioned issue, the researchers have explored several hardware and software-based control strategies. The most widely utilized LVRT enhancement scheme in literature is the Crowbar approach [10, 11]. In this LVRT enhancement approach, a three-phase resistive circuit is placed in series with the CW of the BDFIG to protect the power electronic-based machine side and grid side converters during fault condition without disconnecting the generator from the grid [12]. Despite the very simple operating mechanism and implementation, this approach makes the machine consume a huge amount of power in the resistive circuit during a fault condition and will absorb the reactive power from the grid which consequently leads to unnecessary electromagnetic torque oscillations in the system. Furthermore, the additional resistive circuit requires an extra cost and maintenance which makes it an inefficient and costlier LVRT enhancement option. On the other hand, the software-based LVRT enhancement methods utilize different control schemes to protect the converters during fault conditions [13]. Research studies that have been conducted to determine the effect of control techniques on LVRT capabilities of brushless doubly-fed induction generators include the following. Using an analytical model, Shao et al. [14, 15] introduced an LVRT scheme for symmetric and asymmetric voltage dips based on fixed values for the flux value. [13] is another article that proposes a similar approach to this. According to [14, 15], LVRT can be implemented using positive and negative sequence currents. Other works include [8, 16–18], However, all of these techniques have a similar problem in that the LVRT capability is provided by using fixed reference values for decaying flux. The flux during fault is responsible for the production of electro motive force which in result increase’s fault current and may damage the power electronics converters. Therefore, the flux should be reduced to zero instantly. Moreover, above studies provide poor dynamic response during normal and fault conditions.

To solve the first problem identified above, i.e., use of fixed control flux reference values for reactive current control, in this paper, the reference value for flux is analyzed using an analytical model in the event of a short-circuit fault. The brushless doubly-fed induction generator’s equivalent circuit is used to analyze the load’s reactive current steady state and transient state components. This leads to a dynamic reference value for the control winding current obtained from the load current. Thus, the purpose of this paper is to develop an rotor side converter control scheme for a brushless doubly-fed induction generator that will improve the LVRT characteristics during faults.

All the referred studies have utilized the vector control scheme with manual PI tuning through the “trial and error” method for accomplishing the corresponding control tasks. The quoted PI tuning method is time-consuming and inefficient and does not guarantee the optimal selection of gains at the end of the tuning process; thus, affects the overall performance of the control architecture in terms of managing a suitable dynamics response of the system. To overcome the second stated issue and to enhance the dynamic response of the BDFIG, the authors in reference [19] have utilized an internal model control (IMC) technique to select the optimal PI regulator gains in a vector control scheme. The IMC is an analytical method for extracting the optimal PI gains from the detailed extensive model of the studied control system. Since most of the wind energy conversion systems consists of very complex power and control circuits; therefore, it’s very hard to derive an accurate mathematical of model of such systems. One of the effective methods to overcome this problem is to utilize a PI tuning approach that does not require such extensive modeling of the system and may treat the studied system as black box. One of such approaches is to select the PI gains with metaheuristics-based optimization methods. The operation of these optimization methods is orientated on minimizing or maximizing a pre-defined FF through a stochastic based searching mechanism [20]. A Grouped grey wolf optimizer for doubly-fed induction generator [21], Democratic joint operations algorithm [22], Adaptive fractional-order PID control for Permanent magnet synchronous generator (PMSG) [23] and grouped grey wolf optimizer for a grid-connected PV inverter [24] have been used to achieve maximum power point tracking (MPPT) in mentioned machine controls.

This paper explores the application of one of the most intelligent metaheuristic optimization algorithms called the Salp Swarm Algorithm (SSA) in achieving the optimal dynamic response and LVRT capability enhancement of a grid-tied BDFIG system during normal and fault conditions. The SSA is utilized to select the most optimal PI regulator gains in the proposed vector control scheme. One of the significant aspects of this research work is that the current work utilizes Grid Side Converter (GSC) in conjunction with the Machine Side Converter (MSC) for enhancing the reactive power support to the grid in order to avoid any disconnection during the fault condition. The dynamic response obtained from the SSA-based control scheme is evaluated and compared with the Particle Swarm Optimization (PSO) algorithm-based controller in terms of limiting the fault current, regulating dc bus voltage, and active and reactive power under identical operating conditions. The proposed technique ensures injection of controlled reactive power to the grid during fault conditions to avoid grid disconnection; thus, provides the stable operation of the machine without any usage of hardware such as a crowbar. The major contributions of this article are listed as follows.

An extensive analysis of flux variation during normal and fault condition and developing RSC control scheme.

To improve the dynamic response of the considered WECS under normal and fault operating conditions, the PI regulators’ gains are optimized automatically using the intelligence of SSA.

A comprehensive comparative analysis of the proposed control scheme with IMC and PSO-based VC schemes is made based on few important dynamic response evaluation metrics such as parentage overshoot and settling time to validate the efficacy and essence of the proposed WECS control architecture.

The rest of the paper is organized as follows: Section 2 presents the considered BDFIG and its mathematical foundation. The details of the proposed control scheme are provided in Section 3. Simulation results are discussed in Section 4 while the conclusion for the current research work is laid in Section 5.

2. Modeling of vector control scheme based BDFIG system

This section is further divided into two subsections.

2.1 An analytical model of BDFIG

The BDFIG consists of two windings i.e., Power winding (PW) and Control Winding (CW). The generated power from BDFIG is directly transferred to the grid through PW while the Control winding (CW) is connected to the grid through a power electronic interface [25]. To enable vector control of the machine, the PW and CW voltages are translated into the dq reference frame which is aligned with the PW flux orientation in the current case. The resulting equations for PW, CW and Rotor Windings (RW) are depicted as follows [26]. where V, V and V denote the PW, CW and RW voltage while λ, λ and λ denote the PW, CW and RW flux. Finally, the rotor torque can be calculated as follows;

To control the behavior of the modeled BDFIG based WECS, an SSA-based VC scheme is proposed; the details of which are provided in the subsequent subsection.

2.2 Proposed technique for RSC

The per-phase steady-state equivalent circuit of a BDFIG is shown in Fig 1. The RSC and GSC are decoupled using the capacitance C therefore, the analysis for them can be done independently.

Fig 1

BDFIG equivalent circuit.

To analyses the situation of LVRT during a short circuit fault, the single-phase equivalent circuit for the stand-alone BDFIG system seen from the CW side is shown in Fig 2, i is the a-phase load current, and i represents the short circuit fault current. u and i are the voltage and current for the a-phase at the PCC. Denoting the amplitude and angular frequency of the phase voltage by U and ω, respectively, we can obtain e as follows [5].

Fig 2

Single-phase equivalent circuit from PW.

Taking the Laplace transform of Eq (7), we obtain e in complex frequency domain as follows:

Similarly, for the current source representing the short circuit fault can be written in the frequency domain as: where U is the amplitude of the voltage after the short circuit fault. At natural speed, i = 0 and under normal conditions, it can be assumed without loss of generality that the grid line impedance is negligible. In the presence of a short circuit fault, the switch in Fig 2 will be closed and the stator CW current can be written as:

At natural speed, i = 0 and under normal conditions, it can be assumed without loss of generality that the grid line impedance is negligible. When a load is connected, the switch in Fig 2 will be closed and the stator PW current can be written as: where, the total inductance L, and total resistance R are given by

Defining the magnitude and angle of the total impedance as follows:

Then Eq (11) can be re-written as [27]:

Taking the inverse Laplace transform, we get:

Eq (19) shows that the CW current is composed of two components: first, a fundamental frequency component, and second, an exponentially decaying DC component. The equations for b and c phase currents of the stator CW can be obtained similarly. Thus, the stator CW current consists of a steady state component and transient component. The stator CW current in the synchronous reference frame is obtained using Park transformation: where the steady state and transient state components can be separated as follows:

Thus, the active and reactive CW currents are also composed of a DC steady state component and an exponentially decaying transient component. From above equation flux value is calculated.

2.3 Vector control scheme

The block diagram of the proposed grid-connected BDFIG based vector control scheme is shown in Fig 1.

In the vector control scheme shown in Fig 3, the MSC is directly connected to the CW and is responsible for controlling the active and reactive power with q and d components of CW current respectively [28, 29]. A detailed inside structure of the MSC and GSC are depicted in Figs 4 and 5 respectively.

Fig 3

BDFIG-WECS system.

Fig 4

Machine side converter.

Fig 5

Grid side converter.

As can be seen from Fig 4, the Maximum Power Point Tracking (MPPT) is utilized to extract the maximum available power from the wind energy which is then utilized as the reference active power to be attained from the considered BDFIG based wind turbine. The wind speed Vw is considered as constant (12m/s) throughout the simulation run. To accurately track the reference reactive power with minimum error, four PI regulators with optimal gains are utilized in MSC. The gains of these PI regulators are optimized through SSA to achieve the optimal dynamic response of the system during normal and fault conditions.

Another component of the proposed vector control scheme is GSC as shown in Fig 5. Instantly after the fault occurs, the dc-link voltage overshoots which consequently causes an undesirable operating condition of the BDFIG based power system [16, 30]. To avoid such conditions and to enhance the reactive power support to the grid during fault occurrence, a grid side converter is introduced in the proposed control scheme. The GSC is used to control the dc-link voltage and reactive power whose reference value is set as zero. In typical power grids, a voltage dip may cause a huge deviation in critical system parameters such as active/reactive power, torque, and dc-link voltage. To solve this issue, a coordinated control strategy for the MSC and GSC is explored in current research work. Using a similar approach as that of MSC, the GSC controller is designed with PI tuning through SSA; thus, optimal dynamic response of the proposed system is achieved.

3. Proposed optimization method

This section provides the detailed working mechanism and implementation of the SSA in solving the current optimization problem. Owing to the simple working mechanism, the capability of avoiding stagnation into the local minimum, and good exploration versus exploitation characteristics, the SSA is chosen as the optimization method for selecting the most suitable PI gains in the current study. This is achieved by minimizing an error integrating fitness function through the intelligence of mentioned algorithm. The function opted for the current study is provided in Eq (24). where t is the simulation time while |e1(t)| to |e6(t)| represents the corresponding error signals for the control loops depicted in Figs 4 and 5. The detailed working mechanism of the SSA is provided in the subsequent subsection which is followed by its implementation in the current study.

3.1 Salp swarm optimization algorithm

Salps belong to the family member of the Salpidae group with a translucent barrel-shaped body. To model their movement behavior mathematically, their initial positions are initialized randomly as portrayed in Eq (25) [31]. where; represents the initial positions of the salps with ub and lb as their upper and lower limit respectively while the rand denotes the random values between 0 and 1. Once the search agents (salps) are spread over the pre-defined search space randomly, the swarm is divided into two equal groups i.e., “leaders” and “followers”. This division is based on the evaluation of the pre-defined fitness function. The top half salps in the context of their higher solution quality are considered as leaders while the rest are considered as followers. The solution in evaluating the fitness function for the leading salps is recorded and updated at the end of each iteration using Eq (26). where the symbol shows the jth dimension position of leader salp, symbolizes the jth dimension location of the food source while c1, c, and c3 represents the random numbers. It is important to note that, SSA has the ability to avoid to the local minimum effectively as compared to conventional optimization methods like GA and PSO because of its adaptive optimization nature. The position of follower salps regularly is updated by the SSA and follower salps start to shift gradually toward leading salps. This helps the SSA to prevent stagnating into local optima. Thus, an optimal or close to the optimal solution is produced by SSA during the optimization process. Moreover, SSA performs the best exploration versus exploitation balancing capability. Eq (26) shows that the leader salp location is upgraded with reference to the food source only. The coefficient c1 is the main character in SSA as it contributes to equalizing the exploitation and exploration is calculated by using Eq (27). where l shows the updated number of iteration while L symbolizes the maximum number of iterations. Utilizing Newton’s second law of motion, the upgraded location of follower salp is given as, where i ≥ 2 and symbolize the ith positions of follower salp in jth dimension, t denotes the time while v0 represents the velocity of salps at the initial stage of the optimization process, and is generally considered as 0. As the number of iteration replaces the time during the optimization process and to avoid the fractional number between successive iterations the Eq (29) can be re-written as given underneath. where i ≥ 2 and represents the position of ith follower salp in jth dimension.

3.2 SSA implementation in current optimization problem

The proposed methodology of obtaining optimal PI parameters for MSC and GSC to enhance the dynamic response of the grid-tied BDFIG system using the optimization technique SSA is represented in the flow-chart shown in Fig 6.

Fig 6

Flowchart of the proposed methodology.

A new development in this research is the use of three different optimization techniques (IMC, PSO, and SSA) in order to enhance the dynamic performance of the considered BDFIG based WECS. IMC is a method used to determine PI parameters that considers the operating parameters of the system in order to compensate its open-loop poles with the zeros of the PI controller, which provides the desired closed-loop bandwidth and time constant. A detailed mathematical description of IMC can be found in reference [19]. In contrast, PSO and SSA are swarm intelligence-based optimization techniques that that provide the optimized solution through an iterative process as they use swarm intelligence. As opposed to the IMC, swarm intelligence-based methods do not require the system to have a complete mathematical model. As their very first stage of operation, these optimization methods make use of random number generation within the bounded search space. As a result, the fitness of each search agent is calculated and their position is updated iteratively based on the governing equations of the algorithm until the specified number of iterations has been reached.

Initially, the code for the SSA is encrypted in MATLAB 2018b editor window while the model for the BDFIG along with the proposed control scheme is designed in Simulink. Before starting the optimization process, the optimization parameters such as the number of iterations, numbers of search agents along with their upper and lower bounds, decision variables, and simulation time is encrypted in SSA code. The aim is to achieve optimal values for the decision variables (PI gains in the current study) by minimizing an error integrating fitness function. Like any other metaheuristic-based optimization methods, the SSA initiates its optimization mechanism by spreading a pre-decided number of search agents in bounded search space. Afterward, the fitness function of each search agent is calculated and sorted according to their magnitudes in descending order. During each iteration, the algorithm compares the best among the current search agents with the best among the previously acquired search agents in terms of the magnitude of the evaluated fitness function. If the current value of the fitness function is less than the previously recorded value of the same, the current fitness function value will be stored as global best while if the mentioned condition is not fulfilled, the algorithm will keep the previous solution as the global best. Hence the most optimal PI regulator gains for both the GSC and MSC are achieved at the end of the simulation that corresponds to the least acquired value of the fitness function. Finally, the model is tested with the obtained PI regulator gains and the corresponding dynamic response is recorded. In order to validate the effectiveness of the proposed control scheme, its dynamic response during normal and fault conditions are compared with that of the PSO-based controller under identical operating conditions and system configuration. The optimization parameters utilized for the PSO and SSA are provided in Table 1.

Table 1

Optimization parameters.

Parameters	PSO	SSA
Max. number of iterations (i)	50	50
Population size (N)	50	50
Upper boundary of PI gains (ubj)	10	10
Lower boundary of PI gains (lbj)	0.001	0.001
Number of dimensions (D)	12	12
PSO Cognitive constant (C1)	0–2	-
PSO social constant (C2)	2–0	-
Inertia weight (ɷ)	0.9–0.4	-
Random numbers (C1, C2)	-	Random [0–1]

4. Results and discussion

To evaluate the performance of the proposed SSA-based BDFIG vector control scheme in achieving the enhanced dynamic response and LVRT capability of the mentioned system, the developed model is simulated using MATLAB/SIMULINK version 2018b. The Discrete Tustin/Backward Euler (TBE) with sampling time of 50 μs is used for simulating the developed BDFIG Simulink model. The optimized PI gains achieved at the end of the optimization process at MSC for all the studied methods are provided in Table 2.

Table 2

Optimized PI regulator gains.

Optimization variables	IMC	PSO	SSA
kp1	0.0021	0.0097	0.1301
ki1	0.0152	0.0094	0.0053
kp2	0.1223	0.1902	0.0071
ki2	0.0143	2.0091	0.0051
kp3	0.5455	1.2164	0.8159
ki3	0.8955	1.6219	1.2314
kp4	4.5531	1.2164	2.4044
ki4	0.3401	3.4562	0.9342
kp5	0.6431	1.2431	0.6421
ki5	1.4531	2.8335	1.8550
kp6	0.5690	0.3467	0.3586
ki6	3.2344	1.7443	0.4643

To instigate a fair comparison among the studied methods the proposed BDFIG vector control scheme is simulated with the attained PI regulator parameters under identical operating conditions and system configurations. The machine parameters utilized in the current study are depicted in Table 3.

Table 3

Machine parameters.

Parameters	Value	Parameters	Value
Natural Frequency	52 rad/sec	Power winding resistance (Rsp)	2.3 Ω
PW pole pairs (Pp)	2	Control winding resistance (Rsc)	4 Ω
CW pole pairs Pc	4	Rotor winding resistance (Rr)	1.2967e-4 Ω
PW Voltage	240V	Self-inductance of PW (Lsp)	0.3498 H
CW Voltage	240V	Self-inductance of CW (Lsc)	0.3637H
PW Current	8A	Self-inductance of RW (Lr)	4.4521e-5 H
CW Current	8A	Mutual inductance between PW and RW (Lhp)	0.0031 H
Rated Torque	100 N.m	Mutual inductance between CW and RW (Lhc)	0.0022 H
Grid frequency	50 hz	Rotor moment of inertia (J)	0.4 kg.m2

The effectiveness of the proposed controller is validated under the following performance evaluation cases.

4.1 Convergence profile

In this case, the efficacy of the proposed controller in minimizing the opted FF is evaluated and compared with that of the same for the PSO-based controller under identical optimization parameters and conditions. The convergence curve for both PSO and SSA is depicted in Fig 7.

Fig 7

Convergence curve for PSO and SSA.

It can be observed from Fig 7 that the SSA achieves better solution quality than that of the PSO in terms of minimizing the designed FF. convergence rate and solution quality (minimized FF magnitude) are very important quantities that are responsible to inform directly about the effectiveness of a metaheuristic optimization algorithm. The FF magnitude attained by SSA is 22.00104 while that for the PSO is recorded as 22.23161. The SSA has achieved the stated solution quality in 34 iterations while the PSO has achieved a lesser solution quality as compared to SSA in the 35th iteration of the simulation. Therefore, it is concluded that the SSA achieved a better solution quality and convergence rate than that of PSO under identical optimization parameters and system operating conditions; thus, proves its high-quality optimization capabilities.

4.2 Dynamic response evaluation

To evaluate the performance of the proposed RSC control scheme under abnormal operating conditions, a fault was injected at 2 to 3 seconds of the simulation run. The corresponding voltage of the system is shown in Fig 8.

Fig 8

Voltage waveform during the fault condition.

As can be seen from Fig 8 that the system works under normal conditions for up to 2 seconds of the simulation run. At exactly 2 seconds, a symmetrical fault is injected into the system and as a result, the system voltage drops by 90%. As the fault is removed, the controller starts to restore the rated voltage value as can be observed from 2.5 to 3.1 seconds in Fig 6. Finally, the rated voltage value is regained after the transient period is over; thus, validates the effectiveness of the proposed control scheme. The analytically calculated flux behavior of the machine has been observed in the simulation during the fault condition period. The PW flux without and with controller is shown in Fig 9(a) and 9(b) respectively.

Fig 9

(a) PW Flux without controller (b) PW Flux with controller.

As can be seen from Fig 9(b) that a dip in the PW flux is observed due to the fault which consequently led to a reduction of the PW flux to zero instantly. The role of PW flux is much important in machines during both normal and fault conditions. It causes to produce emf in rotor which in response to produce higher current and makes the machine windings to burn. As opposed to Fig 9(a), it is important to note that despite the occurrence of a fault the PW flux does not enter the critical region and is successfully recovered to the nominal value by the proposed controller. Furthermore, it can also be observed that the PW flux magnitude is decreasing instantly once the fault is injected, thus minimizing the production of the emf which consequently led to a reduction in current which in turn avoids the heating of the machine and connected equipment. Fig 10(a) shows that the PW voltage and current waveforms are in phase with each other during normal conditions, whereas Fig 10(b) shows that the PW current waveform is lagging behind the voltage waveform during fault conditions; thus, provides the required reactive power to the grid.

Fig 10

(a) During normal condition (b) During a fault condition.

In this article, the grid current is controlled to make sure that the machine remains connected with the grid throughout the normal and fault conditions. Fig 11 depicts the efficacy of the proposed control scheme in maintaining the continuity of grid current during the fault condition.

Fig 11

(a) Grid current (b) zoomed version.

4.3 Comparative analysis

Since symmetrical faults in the power grid led to increased machine currents which may cause damage to equipment and loss of capital. For example, an insulated-gate bipolar transistor (IGBT) can withstand 2 per unit CW current for a 1 ms time period. Therefore, it is important to regulate or divert the fault current to prevent any damage to the connected equipment. The proposed control scheme limits the CW current within the specified tolerance levels. In order to validate the effectiveness of the proposed controller, its performance is compared with the PSO and IMC-based controllers under identical operating conditions and system configurations. The corresponding response of the system in terms of d-q current components for the CW and PW is shown in Figs 12 and 13 respectively.

Fig 12

CW Current (a) d-component (b) q-component.

Fig 13

PW Current (a) d-component (b) q-component.

It can be observed from Figs 12 and 13 that the proposed control scheme regulates the CW and PW current components with the least overshoot and settling time; thus, protects the converters from damage due to overcurrent. The proposed controller successfully manages to keep the machine currents within the allowable range even in the presence of a fault at t = 2 seconds. The concerned PW and CW currents return to their pre-fault values after minute sags and swells during the transient period. Besides, it can be observed that the proposed technique results in significantly lowering the oscillations and overshoot in the presence of faults. Tables 4 and 5 show the performance of the SSA in comparison with IMC and PSO for both d and q components of CW and PW currents.

Table 4

CW current values for d and q components.

	Icd			Icq
	IMC	PSO	SSA	IMC	PSO	SSA
tr(s)	0.0055	0.0176	0.00204	0.0024	0.002466	0.002194
ts(s)	1.798	1.799	1.779	1.7992	1.7934	1.434
os(%)	68.654	89.095	64.095	37.209	64.739	41.892
us(%)	6.858	21.5	5.265	1.518	1.974	1.112
pv(A)	144.09	187.61	124.891	175.88	195.136	159.2728

Table 5

PW current values for d and q components.

	Ipd			Ipq
	IMC	PSO	SSA	IMC	PSO	SSA
tr(s)	0.00394	0.0137	0.00274	0.00286	0.0018	0.0021
ts(s)	1.798	1.799	1.786	1.799	1.7797	1.4546
os(%)	67.546	81.51	56.178	31.181	88.950	31.043
us(%)	10.073	14.129	3.741	1.843	1.880	1.600
pv(A)	320.6	421.58	286.68	398.805	444.8337	363.410

In Tables 4 and 5, the tr, ts, os, us, and pv represent the Rise Time, Settling time, Overshoot, Undershoot, and Peak Value respectively. Another very important aspect of any BDFIG based wind turbine control scheme is the regulation of active and reactive power. A comparative analysis of the proposed control scheme with PSO and IMC-based control schemes in terms of the active-reactive power regulation is depicted in Fig 14.

Fig 14

(a) Active Power Response (b) Reactive Power Response.

It is worthwhile to mention that the current research work makes use of an effective control approach to regulate the active and reactive power using PW current. The active power of 235kW is being supplied to the grid through PW during normal conditions as shown in Fig 14(a). During the fault condition, the active power is reduced to almost zero magnitudes and returns to its nominal value without any noticeable oscillations and overshoots once the fault is cleared. The same is observed for reactive power as can be seen in Fig 14(b). The observed dynamic response values for the active and reactive power are shown in Table 6.

Table 6

Active and reactive power.

	Active Power (Pp)			Reactive Power (Qp)
	IMC	PSO	SSA	IMC	PSO	SSA
tr(s)	0.176	0.853	0.210	0.009	0.008	0.006
ts(s)	1.798	1.566	1.798	1.799	1.799	1.788
os(%)	21.208	23.215	11.026	141.71	135.54	103.69
us(%)	22.560	25.560	31.479	1.196	0.915	2.03
pv(w &Var)	3.10e5	3.53e5	2.79e5	1.625e5	2.088e5	1.524e5

One of the most important parameters that need to be regulated is the dc-link voltage. Generally, it is regulated through a grid side converter. In current work, the MSC is utilized to regulate the active power, reactive power, and dc-link voltage through SSA-optimized GSC to achieve the coordination between MSC and GSC. The comparative analysis of the studied control techniques in regulating the dc-bus voltage is depicted in Fig 15.

Fig 15

DC bus voltage response.

Fig 15 shows that the SSA optimization technique performs better during both normal and fault conditions with an improved dynamic response. Following Table 7 shows the corresponding values for the dynamic response for the dc-link voltage.

Table 7

DC link voltage evaluation.

	IMC	PSO	SSA
tr(s)	0.000328	0.000295	0.000263
ts(s)	1.7275	1.685	1.558
os(%)	3.536	2.314	1.183
us(%)	4.126	3.265	2.00
pv(V)	1375	1366	1371

5. Conclusion

This paper presented a flux control technique for the RSC of a BDFIG to inject the required reactive current to the grid by decreasing flux value to zero instantly and restrict the machine from developing electro motive force. The BDFIG is represented by an equivalent circuit and the single-phase equivalent circuit seen from the RSC is used to analyze the steady state components of the flux. A vector control scheme for low voltage ride through under fault conditions in grid-connected BDFIG based wind turbines is presented in this study. Using a crowbar-less approach, the proposed technique improves the dynamic response of the system by optimally tuning PI regulators through the SSA method. During the fault condition, the results shows that RSC provides the required reactive power to the grid by with analytically calculated flux. Simulation outcomes show that the BDFIG successfully rides through fault conditions. To validate the effectiveness of the proposed BDFIG vector control scheme, its dynamic response is evaluated and compared with that of the same with PSO and IMC-based controllers under identical system configuration and operating conditions. The proposed SSA-based BDFIG control scheme provides the most optimal dynamic response among the studied control schemes; thus, validates the effectiveness of the proposed control strategy. The current study has explored the effect of the optimized PI controller with symmetrical faults on the dynamic response of the system. The same optimization method can be explored for asymmetrical faults with different converter compositions. In addition, the nonlinear controllers may be explored and compared with the optimized PI controllers under identical operating conditions and system configurations to suggest the most efficient and suitable controller for the BDFIG based wind turbines.

16 Dec 2021

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Wei Yao, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf.

2. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Comment 1:

Why not directly adopt some advanced nonlinear controllers, but adopt the method of selecting optimizing PI parameters to improve its dynamic performance? The simulation results can be compared with some nonlinear controllers to show the superiority of the proposed control strategy.

Comment 2:

The reviewer think the sentence “Despite the mentioned advantages, the investigation regarding its low voltage ride through (LVRT) is yet explicitly unattended” in the abstract is arbitrary and suggest a revision.

Comment 3:

Note that the format includes paragraphs, references, etc., and some figures are unclear or deformed, such as Fig. 2 and Fig. 5.

Comment 4:

How to get the "8kw" in the following paragraph of Fig. 14?

Comment 5:

Future study should be added in Conclusion.

Reviewer #2: 1. Authors employ meta-heuristic algorithms to optimally tune PI gains of DFIG, such optimization idea and framework is a very typical optimization procedure, however, authors ignored the previous related work, hence, Introduction should be carefully rewritten to cover these important works: (a) Grouped grey wolf optimizer for maximum power point tracking of doubly-fed induction generator based wind turbine, Energy Conversion and Management.2017. (b) Democratic joint operations algorithm for optimal power extraction of PMSG based wind energy conversion system, Energy Conversion and Management. 2018. (c) Energy reshaping based passive fractional-order PID control design and implementation of a grid-connected PV inverter for MPPT using grouped grey wolf optimizer, Solar Energy, 2018. and (d) Adaptive fractional-order PID control of PMSG based wind energy conversion system for MPPT using linear observers, International Transactions on Electrical Energy Systems. 2019.

2. In case studies, more typical meta-heuristic algorithms should be compared with SSA to show its effectiveness and advantages;

3. SSA and PSO parameters should be provided;

4. Simulation environment should be given, e.g., solver type, sampling size, etc.;

5. Other advanced modification of SSA should be reviewed in Introduction, as SSA recently gains a number of interests, authors must address its applications carefully, see: Novel bio-inspired memetic salp swarm algorithm and application to MPPT for PV systems considering partial shading condition, Journal of Cleaner Production. 2019.

6. Section 2 should be significantly reduced as the presented knowledge is well-known, only provide the core contents;

7. In case studies, controller output curve should also be provided;

8. Future studies should be provided.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

17 Jan 2022

Dynamic Response and Low Voltage Ride-Through Enhancement of Brushless Double-Fed Induction Generator Using Salp Swarm Optimization Algorithm

We are grateful to the anonymous reviewers for their comments and suggestions. Below we respond to the comments of each reviewer in detail. We are also providing a revised manuscript that reflects their suggestions and comments. We feel that this has resulted in a stronger manuscript.

Round-01

Response to Reviewer-1

Reviewer-1, Comment -1

Why not directly adopt some advanced nonlinear controllers, but adopt the method of selecting optimizing PI parameters to improve its dynamic performance? The simulation results can be compared with some nonlinear controllers to show the superiority of the proposed control strategy.

Authors’ Response/Clarification

Thank you for this comment. Considering the robust nature and simplicity in modeling and working mechanism, the authors have utilized the PI controller instead of a non-linear controller which is generally complex in design and comparatively sensitive to parameter variations than the PI controller. The major disadvantage of the PI controller which directly impacts the dynamic response of the system is the difficulty in its parameter selection. This research dealt with the mentioned disadvantage of PI controller by optimizing its gains automatically through a swarm intelligence-based metaheuristic technique called salp swarm optimization algorithm. Since the current is more leaned towards the optimization of the PI controller and its impact on the dynamic response of the system so we couldn’t had focused our attention towards the non-linear controllers, nevertheless, owing to this very interesting aspect of the study as mentioned by the reviewer, we have added it in our future recommendations for the researchers to explore it in greater details.

Reviewer-1, Comment -2

The reviewer think the sentence “Despite the mentioned advantages, the investigation regarding its low voltage ride through (LVRT) is yet explicitly unattended” in the abstract is arbitrary and suggest a revision

Authors’ Response

The authors completely agree with the viewpoint of the reviewer and have removed this sentence in the revised version of the manuscript.

Reviewer-1, Comment -3

Note that the format includes paragraphs, references, etc., and some figures are unclear or deformed, such as Fig. 2 and Fig. 5.

Authors’ Response/Clarification

Thank you for this comment. All the figures have been thoroughly checked to improve their quality as suggested.

Reviewer-1, Comment -4

How to get the "8kw" in the following paragraph of Fig. 14?

Authors’ Response/Clarification

Thank you for highlighting this mistake. This typo error has been rectified in the revised version of the manuscript.

Reviewer-1, Comment -5

Future study should be added in Conclusion.

Authors’ Response/Clarification

Thank you for this important suggestion. The following recommendations have been added in the revised version of the manuscript as suggested.

“The current study has explored the effect of the optimized PI controller with symmetrical faults on the dynamic response of the system. The same optimization method can be explored for asymmetrical faults with different converter compositions. In addition, the nonlinear controllers may be explored and compared with the optimized PI controllers under identical operating conditions and system configurations to suggest the most efficient and suitable controller for the BDFIG based wind turbines”.

Round-01

Response to Reviewer-02

Reviewer-2, Comment -1

Authors employ meta-heuristic algorithms to optimally tune PI gains of DFIG, such optimization idea and framework is a very typical optimization procedure, however, authors ignored the previous related work, hence, Introduction should be carefully rewritten to cover these important works: (a) Grouped grey wolf optimizer for maximum power point tracking of doubly-fed induction generator based wind turbine, Energy Conversion and Management.2017. (b) Democratic joint operations algorithm for optimal power extraction of PMSG based wind energy conversion system, Energy Conversion and Management. 2018. (c) Energy reshaping based passive fractional-order PID control design and implementation of a grid-connected PV inverter for MPPT using grouped grey wolf optimizer, Solar Energy, 2018. and (d) Adaptive fractional-order PID control of PMSG based wind energy conversion system for MPPT using linear observers, International Transactions on Electrical Energy Systems. 2019.

Authors’ Response/Clarification

Thank you for this comment. The authors have thoroughly studied these research works and found them very relevant to the current study. Hence all the mentioned studies have been duly added and discussed in the literature review section of the article.

Reviewer-2, Comment -2

In case of studies, more typical meta-heuristic algorithms should be compared with SSA to show its effectiveness and advantages;

Authors’ Response/Clarification

Thank you for this comment. We had compared our results with one of the well-known metaheuristic algorithms called PSO. Since all the metaheuristic algorithms initiate their searching mechanism by generating random numbers, therefore they must be run multiple times to achieve their statistical evaluation. Thus, adding results of other algorithms along with their statistical analysis and comparing them with SSA would add complexity in terms of results interpretation and would also increase the simulation time, paper length, and less important details which may divert the core essence of the article. That’s the reason, the authors have stuck to the major goal of this study which is to explore the intelligence of SSA in achieving an optimal set of PI parameters for enhancing dynamic response and LVRT capability of the BDFIG based wind turbines which is duly achieved as evident from the outcomes of the study.

Reviewer-2, Comment -3

SSA and PSO parameters should be provided;

Authors’ Response/Clarification

Thank you for this comment. The utilized parameters for the PSO and SSA are added in tabular form (Table 1) in the revised version of the manuscript as suggested.

Reviewer-2, Comment -4

Simulation environment should be given, e.g., solver type, sampling size, etc.;

Authors’ Response/Clarification

Thank you for this comment. The “Discrete Tustin/Backward Euler (TBE)” solver is used for simulating the proposed BDFIG based wind turbine model with a sampling time of 50 µs. This information regarding the solver has been added and highlighted in the revised version of the manuscript as suggested.

Reviewer-2, Comment -5

Other advanced modifications of SSA should be reviewed in the Introduction, as SSA recently gains a number of interests, authors must address its applications carefully, see: Novel bio-inspired memetic salp swarm algorithm and application to MPPT for PV systems considering partial shading condition, Journal of Cleaner Production. 2019.

Authors’ Response/Clarification

We gladly take this suggestion. The mentioned study is very relevant to the current research work and hence added in the revised version of the manuscript as suggested.

Reviewer-2, Comment -6

Section 2 should be significantly reduced as the presented knowledge is well-known, only provide the core contents;

Authors’ Response/Clarification

We gladly take this suggestion. The mentioned Section has been reduced accordingly.

Reviewer-2, Comment -7

In case studies, controller output curve should also be provided;

Authors’ Response/Clarification

Thank you for this comment. The controller output in the current study is controlled pulses used to trigger the inverter. Since these pulses have been found meaningless thus their interpretation in the text would not be possible, therefore, the authors have not added them in the article.

Reviewer-2, Comment -8

Future studies should be provided

Authors’ Response/Clarification

Thank you for this important suggestion. The following recommendations have been added in the revised version of the manuscript as suggested.

“The current study has explored the effect of the optimized PI controller with symmetrical faults on the dynamic response of the system. The same optimization method can be explored for asymmetrical faults with different converter compositions. In addition, the nonlinear controllers may be explored and compared with the optimized PI controllers under identical operating conditions and system configurations to suggest the most efficient and suitable controller for the BDFIG based wind turbines.

Submitted filename: PLAS ONE reviewres comments(1).docx

Click here for additional data file.

7 Mar 2022

Dynamic Response and Low Voltage Ride-Through Enhancement of Brushless Double-Fed Induction Generator Using Salp Swarm Optimization Algorithm

PONE-D-21-35834R1

Dear Dr. MEMON,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Wei Yao, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

Reviewer #2: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

20 Apr 2022

PONE-D-21-35834R1

Dynamic Response and Low Voltage Ride-Through Enhancement of Brushless Double-Fed Induction Generator Using Salp Swarm Optimization Algorithm.

Dear Dr. Memon:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Professor Wei Yao

Academic Editor

PLOS ONE