A comprehensive review of wind power based power system frequency regulation

Abstract

Wind power (WP) is considered as one of the main renewable energy sources (RESs) for future low-carbon and high-cost-efficient power system. However, its low inertia characteristic may threaten the system frequency stability of the power system with a high penetration of WP generation. Thus, the capability of WP participating in the system frequency regulation has become a research hotspot. In this paper, the impact of WP on power system frequency stability is initially presented. In addition, various existing control strategies of WP participating in frequency regulation are reviewed from the wind turbine (WT) level to the wind farm (WF) level, and their performances are compared in terms of operating principles and practical applications. The pros and cons of each control strategy are also discussed. Moreover, the WP combing with energy storage system (ESS) for system frequency regulation is explored. Furthermore, the prospects, future challenges, and solutions of WP participating in power system frequency regulation are summarized.

Graphical abstract

Keywords

frequency regulation strategies / wind turbine generators / grid-forming control / model predictive control / energy storage system

1 Introduction

Due to the over-reliance on traditional fossil fuels, global warming is becoming intensively severe, leading to the frequent occurrence of extreme weather events around the world [1]. Aware of the global climate crisis, 165 countries have set renewable energy targets, including investing heavily in the energy structural transition toward net-zero carbon emission and targeting 100% renewable energy [2]. China also unveiled the carbon peaking and carbon neutrality goals before 2030 and 2060, respectively [3,4].

As the most competitive and fastest growing renewable energy source (RES), wind power (WP) was vigorously developed and widely utilized in the world in the last few decades. By the end of 2021, the cumulative installed capacity of global WP reached 837 GW, where China led the ranking of the largest WP countries worldwide with a total installed capacity of 338.31 GW and accounted for 40.40% of the global share [5]. The total installed capacity of WP in the United States was 134.40 GW, accounting for 16.05%, second only to China. The cumulative installed capacity of WP in China and the United States occupied more than 50% of the global total installed capacity, while the top five countries, including Germany, India, and the UK, accounted for more than 70% [6]. However, wind turbines (WTs) are connected to the grid through power electronic converters, which cannot sense the frequency disturbance of the power system [7]. Thus, the growing penetration of WP greatly weakens the system inertia and further threatens the frequency stability of the power system [8,9].

With the increasing penetration of WP in the power system, many frequency safety accidents of the power system have occurred in recent years. For example, on August 9th, 2019, a major blackout occurred in the Great Britain power system. Before the accident, more than 30% of the load was provided by the WP output. A single line-to-ground fault was caused by a lightning strike, which indirectly led to the power output of the Hornsea offshore WF plummeted by 737 MW. The cumulative active power loss of the system reached 1131 MW, which caused the rate of change of frequency (RoCoF) to exceed the protection threshold value of 0.125 Hz/s and further led to the tripping of 350 MW distributed energy resources [10]. On September 28th, 2016, typhoons and torrential rains hit the power grid in southern Australia, leading to a major system blackout for more than 50 h, where the penetration of RES was 48.36% with 14 wind farms and 5 synchronous generators connected to the grid [11]. From the system event compliance report [12], the insufficient system inertia, which was at an all-time low of 3000 MW·s before the accident, was one of the main reasons for the subsequent system collapse.

Therefore, it is vital for WP to actively participate in frequency regulation to improve the overall resilience of the power system during severe weather, and many countries have revised their grid codes and standards. Fig.1 shows the comparison of grid codes for WP frequency regulation from several countries [13–21]. The technical specification for connecting WF to the power system of China requires that all WFs should have the inertia response and primary frequency regulation (PFR) capability, among which the dead zone of PFR should be set between 0.03 to 0.1 Hz and the primary frequency settling time should be less than 15 s [20]. The technical requirements in Hydro-Quebec of Canada require WFs with an installed capacity greater than 10 MW to provide frequency modulation power of more than 5% of the rated capacity for more than 10 s in the event of a rapid frequency change with a frequency deviation greater than 0.5 Hz and a duration of less than 10 s, which can provide similar control effect of conventional generators with an inertia time constant of 6 s [21].

There are a number of review papers on the topic of WP based power system frequency regulation from different perspectives. In Ref. [22], selected European grid code requirements, including Germany, Spain, Ireland, and UK, are presented. The need for WP to participate in PFR, secondary frequency regulation (SFR), and tertiary reserves are discussed. Tang et al. [23] reviewed the WP participating in PFR in the aspects of control methods and capability assessment from WT level and wind farm (WF) level. In Ref. [24], control methods proposed for both WP and solar PV generators in PFR were reported. A more detailed elaboration of the inertia response of WP was illustrated in Ref. [25], together with the performance comparison of different types of WTs. The role of energy storage systems (ESS), including distributed and aggregated configurations in the WF, was described in Ref. [26], which proposed ESS as an alternative for de-loading to provide PFR. Ye et al. [27] discussed the interaction of WP and other traditional fossil-based power sources, which introduced an improved automatic generation control during SFR. The secondary frequency drop (SFD) that occurred in the frequency regulation process was emphasized and the control methods to arrest the SFD were assessed in Ref. [28]. While the available review papers lay a good foundation for the principles and control methods of WP participating in system frequency regulation, this paper assessed the spatio-temporal frequency distribution as the increasing penetration of converter-based WP, where the widely adopted concept of center of inertia (COI) was no longer suitable. In addition, it discussed the emerging application of grid-forming control and believed that it would be the future development trend of WP in system frequency regulation. Moreover, as an updated review of different state-of-art control strategies provided by WP at both the WT level and the WF level, it evaluated the pros and cons of each control strategy. Furthermore, it addressed the prospects, future challenges, and possible solutions concerning power system frequency regulation with the increasing contribution of WP.

2 Impact of WP on power system frequency

Compared with conventional power generation sources, such as fossil fuel power plants, the WP output is closely related to the random change of wind speed. When the penetration of WP increases, the fluctuation and randomness of WP will inevitably affect the frequency stability and the dynamic characteristic of the power system, specifically reducing the system inertia and PFR capability.

The conventional power system frequency response is depicted in Fig.2, which can be divided into three stages, inertia response, PFR, and SFR [29,30].

Inertia response is the natural response process of the synchronous generators to sudden changes in system frequency, which generally lasts within 10 s. The active power support provided during the inertial response stage comes from the release of the rotational kinetic energy stored by each synchronous generator and the load equipment, which depends on the system inertia and the unbalanced power as expressed in Eq. (1) [29].

(1)\frac{\mathrm{d}f}{\mathrm{d}t}=\frac{f_0}{2H_{\mathrm{S}}{S}_{\mathrm{B}}}(P_{\mathrm{m}}- P_{\mathrm{e}}),

where df/dt, f₀, H_S, S_B, P_m, and P_e are the RoCoF, system frequency, total system inertia, rating power of the generator, mechanical power, and electromagnetic power, respectively.

When the system frequency changes beyond the threshold value, the PFR control will be activated to adjust the governor system of the generator. The P_m and P_e are rebalanced during PFR and further frequency change is prevented. PFR usually lasts 10 to 30 s and a frequency deviation (Δf) still exists, as shown in Fig.2. After 30 s, the SFR control will be activated by adjusting the output power reference of generators to return the system frequency to its nominal value, which usually lasts 30 s to 30 min [31,32].

As the penetration of WP continues to increase and gradually replaces conventional power sources in the system, its frequency response characteristics, which are significantly different from synchronous generators, will undoubtedly affect the frequency dynamic behavior of the system. References [33–36] analyzed the influence of WP fluctuations on power system frequency from the perspective of frequency deviation, frequency nadir, and RoCoF, which illustrate that the increasing penetration of WP will weaken the inertia capability of the system, deteriorate the frequency dynamics of the system, and threaten the safe and stable operation of the system.

Additionally, with the increasing share of RES and power electronics in power systems, the differentiation of the spatio-temporal frequency distribution emerges [37]. At the moment of disturbance in the power system, the system power shortage will be distributed among the synchronous generators according to the electrical distances between the nodes of synchronous generators and the disturbance point. This distribution will cause the deviation between the rotational speeds of different synchronous generators, resulting in the difference in the frequency and the voltage phase angle between the nodes, which further affects the power flow distribution in the system and redistributes the power shortage of the system until the frequency of each node reaches synchronization again [38]. In the traditional power system, although the differentiation of the spatio-temporal frequency distribution exists objectively, its impact on the system is relatively small. The concept of COI was proposed in Ref. [39] and the system frequency response model was commonly adopted to depict the power system frequency dynamics [40]. A schematic diagram of COI and a classic system frequency response model are demonstrated in Fig.3 [41,42].

However, as the system inertia keeps decreasing due to the increasing share of RES, the frequency fluctuation is more severe under the disturbance. The uneven distribution of the unbalanced power during the power angle swing process may lead to a significant difference in the RoCoF of each node at the early stage of the disturbance, which could unintentionally trigger the protection of power electronic equipment. Li et al. [43] pointed out that the observed frequency response synchronization delays were considerable, even reaching 2.3375 s.

Therefore, it is necessary to make WP effectively participate in the power system frequency regulation and massive research has been conducted in this area.

3 Conventional operation and control principle of variable-speed WTs (VSWTs)

Fig.4 exhibits the schematic diagrams of the main types of WTs that are currently in commercial use and operation, i.e., squirrel cage induction generator (SCIG) [44,45], doubly fed induction generator (DFIG) [46], and permanent magnet synchronous generator (PMSG) [47].

SCIG is directly connected to the power system through soft starters, which is also categorized as fixed-speed WT with narrow wind speed ranges. The fixed-speed WTs have a certain frequency support capacity, but they can only provide a relatively small active power support when the system frequency disturbance occurs. DFIG and PMSG, categorized as VSWTs, are connected to the power system via power electronic converters, which help the WTs work in a variable speed constant frequency mode to regulate the output power over a wider range of wind speeds. Meanwhile, VSWTs can control both active power and reactive power independently, which own a better voltage support and fault ride-through capabilities [48–50]. However, VSWT cannot actively support the system frequency since its power electronic converter connection decouples the rotor speed and the power system frequency changes [51,52], which poses challenges to the system frequency stability.

Fig.5 shows the typical control system structure of a DFIG [23,53], which includes three control subsystems, a WT control, a rotor side converter (RSC) control, and a grid side converter (GSC) control. For the WT control, the pitch angle controller, the maximum power point tracking (MPPT) controller, and the speed controller take charge of the energy conversion from WP to mechanical power. For the RSC control, its main goal is to control the electromagnetic power of the DFIG to keep the rotor always running at the optimal speed to achieve the MPPT. It obtains the reference power values (P*) through the speed controller and uses the rotor d-axis current (I_d*) controller to ensure that the actual power is equal to its reference value. For the GSC control, it not only facilitates power exchange with the RSC, but also provides an additional reactive power support for the power system. For the vector control-based DFIG, a phase-locked loop (PLL) is utilized to capture the frequency and phase of terminal voltage (θ_s) to synchronize with the power system, which is commonly adopted and categorized as grid-following (GFL) control. Meanwhile, a new concept of grid-forming (GFM) control is proposed to enable synchronization without PLL [24], which has a better stability when connecting to weak power grids. However, the existing studies on GFM are mainly applied in micro-grids or islanded power systems and its application in active power grids is still worthy of further research [54].

WT control aforementioned is the primary process of energy conversion in WP generation, which is used to intercept the kinetic energy carried by the flowing air and convert part of it into mechanical energy. Based on aerodynamics, the captured mechanical output power of WT can be calculated as [55]

(2)P_{\text{m}} =\frac{1}{2}\rho A{v}_{\text{w}}^3{C}_{\text{P}}(\lambda ,\beta ),

where P_m, ρ, A, v_w, and C_P is the captured mechanical power, air density, swept area by the blade, wind speed, and power coefficient, respectively. Besides, C_P is a function of pitch angle (β) and tip speed ratio (λ), which can be expressed as

(3)\lambda =\frac{{\omega }_{\mathrm{r}}R}{v_{\text{w}}}.

The characteristic of C_P is provided by the manufacturer, and this characteristic is different from the WTs produced by different manufacturers. Based on the Betz limit, the maximum value of C_P is 0.593; λ is related to v_w, the radius of the rotating blade (R), and the rotor speed ({\omega }_{\mathrm{r}}).

Fig.6 manifests the mechanical power characteristic curve of DFIG as a function of {\omega }_{\mathrm{r}} and β [56]. Based on the aforementioned pitch angle controller and MPPT controller, there is an optimal C_Popt that enables the WT to maintain the optimal power output (P_opt) at various wind speeds. In Fig.6, the red line connecting the P_opt is called the MPPT curve, and the P_opt can be expressed as

(4)

(5)k_{\text{opt}}=\rho \text{π}{R}^5\frac{C_{\text{Popt}}}{2{\lambda }_{\text{opt}}^3},

where k_opt is the constant for tracking the MPPT curve, which can be obtained by Eq. (5); λ_opt is the optimal tip speed ratio; {\omega }_0 is the cut-in angular speed, and 0 < {\omega }_{\mathrm{r}} < {\omega }_0 corresponds to the starting zone; {\omega }_0≤ {\omega }_{\mathrm{r}}≤ {\omega }_1 corresponds to the MPPT zone, also known as the optimization zone; {\omega }_1 < {\omega }_{\mathrm{r}}≤ {\omega }_{max} corresponds to the quasi-constant speed zone, where a droop characteristic curve is utilized to avoid the sudden output change near the maximum angular speed {\omega }_{max}; in the constant power zone, the pitch angle controller will be activated to keep the output at the rated value (P_max).

4 Frequency regulation strategies for VSWTs

From the above-mentioned analysis, the active power output of the VSWTs in the conventional MPPT mode is only subject to wind speed change and the active power controller cannot sense the system frequency deviation. In that case, the conventional operation of VSWTs can neither contribute to the system inertia response nor participate in the system PFR, which is contrary to the grid code requirements and the development of future grid-friendly RES. Therefore, an extensive amount of research has been conducted on the frequency control strategies for VSWTs [57–63]. The classification of control strategies for VSWTs is depicted in Fig.7, which can be classified from the aspects of the physical nature of the control principles and the converter control architecture.

4.1 De-loading control

As described in Section 3, the WT normally operates in the MPPT mode to maximize the utilization of wind energy, and cannot actively participate in the system frequency regulation. The de-loading control changes the operating point of the WT to deviate from the MPPT curve and to run at the sub-optimal power point, leaving an active power reserve as backup, which is also known as the power reserve control. At present, there are two types of de-loading control strategies, pitch angle control and overspeed control.

4.1.1 Pitch angle control

According to Eqs. (2) and (3), C_P and P_m are positively correlated, where C_P is a function of β and λ that correlates to {\omega }_{\mathrm{r}}. Therefore, the adjustment of {\omega }_{\mathrm{r}} and β can both change the output power of the WT. Fig.8 presents the control block diagram of pitch angle control, where Δβ₀ is the increased pitch angle in steady-state, k% is the de-loading coefficient, β_ref is the pitch angle reference value, (dβ/dt)_max and (dβ/dt)_min are the pitch rate limit, and β_max is the maximum margin for the variation of pitch angle. In the partial load region, β_ref is fixed at zero degrees to track the MPPT. Meanwhile, the pitch angle control increases Δβ₀ in advance to reduce the wind energy capture efficiency, and therefore the WT runs below the MPPT curve reserving a certain amount of active power for PFR [64,65]. In the full load region, β_ref increases following the change of wind speed to keep the WT at rated power. When the system frequency decreases, the pitch angle decreases, releasing the captured mechanical power to provide additional active power support. Fig.9 shows the power−rotor speed curve of WT under pitch angle control [66].

In Fig.9, the red dotted line is the MPPT curve for the WT in normal operation, and the blue dotted line is the de-loading curve with pitch angle control. By adjusting the pitch angle in advance, the active power output in the de-loading curve can differentiate from the MPPT curve by a fixed amount of ΔP as power reserve backup at different wind speeds, ensuring that the WT can provide active power support under various operating conditions.

However, the response of pitch angle control is relatively slow considering the involved mechanical parts. Meanwhile, the excessive use of pitch angle control will increase the mechanical tension of the WT and thus shorten its service life [67,68]. Besides, the WTs working in the de-loading curve during steady-state operation to obtain the power reserve backup harm the economic benefits in the long run.

4.1.2 Overspeed control

Different from the control principle of pitch angle control, overspeed control achieves the effect of de-loading by changing the rotor speed ({\omega }_{\mathrm{r}}), which is accomplished by torque control following a de-loaded sub-optimal power point tracking curve. The operating point of WT can be shifted left by decreasing the {\omega }_{\mathrm{r}} or shifted right by increasing the {\omega }_{\mathrm{r}}, and either way can obtain the power reserve during the steady state. However, the authors of Refs. [69,70] proposed that the operating point on the left side might cause small signal instability. Thus, the overspeed control is preferable. The active power reference for the overspeed control is expressed as

(6)P_{\text{de-ref}}=P_{\text{opt}}( 1-k\mathrm{\%}),

where k% is the de-loading coefficient, the same as k% in pitch angle control. For the de-loading power reference, a corresponding sub-optimal power coefficient C_Pde can be given as

(7)C_{\mathrm{P}\mathrm{d}\mathrm{e}}=\frac{P_{\mathrm{o}\mathrm{p}\mathrm{t}}(1-k\mathrm{\%})}{0.5\rho A{v}_{\mathrm{w}}^3}=C_{\mathrm{P}\mathrm{d}\mathrm{e}}(v_{\mathrm{w}}, {\omega }_{\mathrm{r}},\beta ).

In Eqs. (6) and (7), for a given wind speed and pitch angle, the change of {\omega }_{\mathrm{r}} can be accomplished based on the pre-defined de-loading coefficient. In Fig.10, for overspeed control, the operating point is shifted to the right from point A to point C by increasing the {\omega }_{\mathrm{r}}, which leaves a certain amount of active power as backup. During frequency response, the operating point will shift from point C back to point B, which is the new equilibrium point and the reserved power will be released to support the system frequency [71,72].

Li & Zhu [73] analyzes the PFR capability of overspeed control quantitatively, and points out that the response of overspeed control is much faster than that of the pitch angle control. But its application is limited by the maximum rotor speed of WT and is preferably used at low and medium wind speeds. Considering the rotor speed constraint of overspeed control, Mu et al. [74] studied the PFR performance of DFIG in the case of long-period continuous frequency disturbances, and proposed an improved overspeed control scheme with bi-directional PFR capability. Wang & Shao [75] reported the long-term field operations of multiple WFs located in mountainous terrains of Yunnan, China, indicating that overspeed control can provide reliable PFR for the power system.

The de-loading control discussed above requires the WT to deviate from the MPPT curve in normal operation, which reduces the wind energy capture efficiency. However, power system frequency regulations occur occasionally. To cope with small probability events, the long-term operation of WTs in a sub-optimal state will cause a significant amount of power loss and sacrifice the economic benefits. Therefore, rotor kinetic energy control is proposed to keep the WTs in the MPPT mode under normal operating conditions.

4.2 Rotor kinetic energy control

The principle of rotor kinetic energy control is to introduce frequency-related control links into the active power control system of the WT by releasing the stored kinetic energy in the rotor during frequency response [76,77]. At present, there are three types of rotor kinetic energy control strategies, short-term overproduction control, droop control, and virtual inertia control.

4.2.1 Short-term overproduction control

Fig.11 shows the rotor kinetic energy releasing process of WT and its power characteristic curve during short-term overproduction control at a wind speed of 0.6 p.u. (MPPT zone) and 0.93 p.u. (quasi-constant speed zone), which illustrates that the control process of the rotor kinetic energy release in WT consists of two stages, a power support stage and a speed recovery stage [78,79]. When the system frequency decreases, the WT can provide an active power boost (ΔP_boost) by increasing the electromagnetic power instantaneously. In Fig.11(a), under the 0.6 p.u. wind speed condition, the electromagnetic power change process follows trajectory A-B-C, with the rotor speed ({\omega }_{\mathrm{r}}) dropping sharply in the support stage, and the mechanical power changes from A to D. Additionally, a critical situation is given as well, where power support stage lasts long enough and {\omega }_{\mathrm{r}} reaches the lower limit at {\omega }_{min} (point S). In that case, the electromagnetic power change process follows trajectory A-B-C-S. In the speed recovery stage, to restore {\omega }_{\mathrm{r}} to its initial value, the given value of electromagnetic power is less than the current captured mechanical power, and therefore the change of electromagnetic power follows trajectory C-E-A. In the meantime, the mechanical power gradually increases from D to A as {\omega }_{\mathrm{r}} increases. Finally, the electromagnetic power and the mechanical power reach the initial equilibrium point A. For the 0.93 p.u. wind speed condition, the same analysis can be conducted as well, with trajectory A’-B’-C’-D’-E’-F’-A’. When {\omega }_{\mathrm{r}} reaches the upper limit in the constant speed zone, the converter in the WT can increase the active power output through its short-time overload capability to participate in the short-term overproduction control as well [80].

In Fig.11(b), the shaping parameters (ΔP_boost, t₁, t_drop, ΔP_recovery, t₂, and t_recovery) need to be optimized to give full play to the PFR of short-term overproduction control. In Ref. [81], those parameters were optimized based on the particle swarm optimization algorithm, which could greatly reduce the negative effect during speed recovery stage, and the quantitative analysis showed that the proposed method could meet the stringent grid code requirement of PFR.

4.2.2 Droop control

Droop control, also known as proportional control or slope control, mimics the droop characteristic curve of synchronous generators during PFR [82,83], given in Fig.12(a). Fig.12(b) shows the control block diagram of droop control, where the active power support (∆P₁) is proportional to the frequency deviation (Δf) and droop gain (R), and LPF is the abbreviation of the low-pass filter.

The rotor kinetic energy contained is closely related to the wind speed, and the use of a unified droop coefficient cannot fully liberate the potential. When the wind speed is relatively low, the rotor kinetic energy of WTs is limited, and excessive use of it can easily lead to PFR end prematurely. Therefore, the authors of Refs. [84,85] introduced a variable droop control strategy by continuously adjusting the droop coefficient with changes in the wind speed. However, the power system was highly nonlinear and a linear droop coefficient could lead to system stability problems [86–88]. Huang et al. [89] analyzed the small-signal stability of conventional P−f droop control and reported that the rotor speed might exceed the admissible range during PFR, where WTs would be out of synchronization. A modified synchronized control was proposed by adding an assistant damping component, which improved both the synchronous stability and the small-signal stability of the existing droop control. Arani & Mohamed [90] proposed an efficiency droop control for different wind speed conditions, and both transient and steady-state analyses of frequency dynamics were performed.

4.2.3 Virtual inertia control

Similar to droop control, virtual inertia control is to simulate the inertia response characteristics of conventional synchronous generators [91,92]. The relationship between the inertia constant and rotor speed for a synchronous generator can be expressed as

(8)2H_{\mathrm{S}}\frac{\mathrm{d}\mathrm{\Delta }\omega }{\mathrm{d}t} =T_{\mathrm{M}}-T_{\mathrm{E}}-D\mathrm{\Delta }\omega ,

where H_s is the inertia constant, \mathrm{\Delta }\omega is the change of generator speed, T_M is the mechanical torque, T_E is the electromagnetic torque, and D is the damping coefficient. In Eq. (8), when system frequency changes and neglects the change of T_M, T_E will change accordingly. To emulate this characteristic, an additional control loop is added to the WT [93]. Fig.13 shows the control block diagram of virtual inertia control, where the active power support (ΔP₂) is proportional to the RoCoF (df/dt) and droop gain (K_f), and LPF is the abbreviation of the low-pass filter.

The rotor speed of VSWT has a large range of variation (usually from 0.7 to 1.2 p.u. for DFIG and 0.5 to 1.2 p.u. for PMSG). Therefore, it can provide sufficient active support at the initial stage of frequency disturbance, which can effectively arrest the drop of RoCoF and raise the frequency nadir [94,95]. Even so, virtual inertia control can only last for a short period of time (generally less than 6 s), as the rotor speed hits its minimum limit. In the rotor speed recovery stage that follows, the rotor speed of the WT will return to the MPPT curve, and the acceleration of rotor speed will decrease the active power output, which may lead to a SFD in the system [96,97]. To avoid this problem, the WT should be kept running stably at a lower rotor speed after releasing the kinetic energy of the rotor, and the rotor speed should be restored after the system frequency is stabilized [98,99]. References [100,101] reported that the application of inertia control would cause torsional oscillation in the drivetrain of DFIG, which might cause unexpected shutdown and will greatly reduce the service life of mechanical components. In Ref. [102], a damping control strategy was proposed based on the modulation of the reactive current in GSC to suppress torsional oscillation.

4.3 GFM control

All of the abovementioned rotor kinetic energy controls, also categorized as GFL control, rely on the PLL to synchronize with the grid, which support system frequency regulation by adding supplementary controls in the power control loop of VSWTs. The main disadvantage of the GFL control is that the activation of frequency regulation requires that the input signal is measured from the PLL, which contains inevitable time delay, such as the LPF in the droop control and virtual inertia control. Different from the GFL control, the recent emerging GFM control mimics the behavior of synchronous generators to act as a controllable voltage source with a constant output voltage and frequency [103,104], which can provide a simultaneous inertia response during the system frequency disturbance. Fig.14 graphically explains the differences between GFL control and GFM control [105].

An enormous amount of research has been conducted on GFM control from its synchronization stability, fault ride-through capabilities, voltage support capabilities, inertia response, and frequency response capabilities, etc. [106–108]. Several control strategies of GFM have been proposed in recent years as well, including droop control [109], power synchronization control [110], matching control [111], and virtual oscillator control (VOC) [112,113]. Droop control emulates the speed droop control of the synchronous generator, which contains a simple structure and is widely implemented in micro-grids, and is regarded as the baseline solution of GFM control [109]. Power synchronization control is initially proposed for high-voltage DC (HVDC) applications in Ref. [110], which overcomes the instability of PLL-based connection in a weak grid. Matching control exploits the similarities between the DC-link voltage of the converter and rotor angular frequency of the synchronous generator, which does not rely on any measurements from the AC-side and greatly reduces the corresponding time delay [111]. Different from all the previously mentioned strategies, VOC mimics the dynamic of the weakly nonlinear limit-cycle oscillator, which can globally synchronize a converter-based power system from an arbitrary initial condition [112]. A fully dispatchable VOC is further developed in Ref. [113].

The inertia response and PFR performances of GFM-based controls have also been investigated in both DFIG and PMSG, respectively [114–119]. Xi et al. [114] proposed that other control functions, e.g., MPPT and DC-link bus voltage control, would have dynamic coupling with the GFM control, which might affect the equivalent inertia constant and damping coefficient of the PMSG and thus weaken its frequency response capability. For PMSG, both the machine side converter (MSC) and the grid side converter (GSC) can obtain the MPPT control. However, the energy utilized to realize the frequency response is different for those two types of implementations. The DC-link bus capacitor will be the energy source for frequency response if MPPT is realized by the MSC, and the rotor will be the energy source if MPPT is realized by the GSC. Fig.15 shows the difference between those two control strategies [114]. It is worth noting that the energy provided by the DC-link bus capacitor is comparatively small compared to that of the rotor.

However, GFM control suffers from the small-signal instability when connecting to a stiff AC grid [120], and the current-limiting issues during the fault-ride through may cause transient instability [121,122]. Moreover, the current research on GFM mainly focuses on the design of synchronization control strategies as introduced above, and further inner and outer control loops (e.g., for frequency and voltage support) need to be integrated [123].

4.4 Coordinated control

In summary, the frequency regulation control strategies for VSWTs are discussed in Sections 4.1–4.4, and their relative benefits and drawbacks are concluded in Tab.1.

From Tab.1, it can be observed that the de-loading control, including the pitch angle control and overspeed control, can release more energy during frequency regulation. In Ref. [124], considering that the overspeed control had the characteristics of faster response and smaller mechanical wear, it was preferentially used during PFR. Only when the overspeed control is not applicable due to the upper limit of rotor speed, the pitch angle control is used. However, the sub-optimal operation of de-loading control during normal conditions sacrifices the economic benefits of WF. Short-term overproduction control and virtual inertia control can effectively arrest the RoCoF decline during inertia response, but the excessive release of kinetic energy may cause serious SFD during the rotor speed recovery process. References [125,126] combined the droop control and virtual inertia control, and utilized a fuzzy adaptive controller to adjust the droop coefficient based on wind speed and frequency changes, which greatly improved the PFR performance and avoided the SFD. References [127–133] proposed the coordinated control strategy of de-loading control and rotor kinetic energy control, where a sub-optimal power point tracking was introduced to enhance the fast response with more kinetic energy stored, which could also mitigate SFD risks.

The emerging GFM control for power converter has a faster inertia response capability with its intrinsic frequency response characteristics similar to those of the synchronous generators. The capability of operation in weak grid or even islanded grid of GFM control plays its role as the keystone for future energy transient toward 100% RES possible. Although the current GFM control suffers from transient stability issues and current-limiting issues, continuing research has proposed potential solutions, such as the virtual impedance control [134,135]. In addition, the coordinated implementation with GFL and GFM [136,137], which considers the share ratio in real-time application to archive their complementary benefits, is worthy of further research.

5 Frequency regulation strategies at the WF level

The direction and magnitude of wind speed in nature are affected by various random and uncontrollable factors, such as atmospheric pressure, temperature, and topography [138,139]. Usually, a WF contains hundreds of WTs which are located on different terrains and at different heights. During the actual operation, the performance of each WT may be affected by turbulence [140], wind shear effect [141,142], tower shadow effect [143,144], and wake effect [145,146], which will result in significantly different output power for each WTs in the same WF. However, the conventional way simply treats the WF as an aggregated model with identical WTs, which ignores the different operating statuses of WTs in actual operation. During frequency response, the aggregated model of WF will bring stability problems, and reasonable coordination within the WF, is the focus of extensive research.

Based on the control topology, the frequency regulation strategies at the WF level can be divided into distributed control and centralized control [147]. For distributed control, the WTs in the WF participate in system frequency spontaneously according to the pre-defined control strategies as covered in Section 4. For centralized control, the active power operating point of each WT is commanded by the WF control center during frequency regulation and communications are necessary. The schematic response time series of distributed control and centralized control are shown in Fig.16.

5.1 Distributed control

Considering the different operating statuses of WTs in the WF, the uniform control strategy that allocates active power evenly for each WTs during the frequency response is clearly unreasonable [148]. References [149,150] utilized a variable parameter proportional to the wind speed, which ensured that the WT with a high wind speed could release more kinetic energy and vice versa. Similarly, a weighting factor that divides the WTs based on wind speed segment was deployed in Ref. [151], which reduced the measurement error of wind speed. In Ref. [152], a combination of virtual inertia control and de-loading control was proposed with wake effect considered, which indicated the utilization of pitch angle control in the upstream WTs could effectively reduce the wake interactions. The fuzzy C-means clustering algorithm was applied to develop the equivalent WF model considering the wake effect [153].

For WF that widely uses rotor kinetic energy control as PFR strategy, when a large number of WTs enter the speed recovery stage at the same time, a large active power deficit will arise, which will lead to a serious SFD [154]. A time-varying inertia and droop controls were proposed in Ref. [155], where the control gains are related to the frequency-response time to mitigate the SFD. In Ref. [156], the adaptive power compensation control was introduced where online system disturbance estimation was conducted by utilizing the power excitation method, which adjusted the control strategies based on the real-time frequency deviation and RoCoF. The power increment and duration time of short-term overproduction control were modeled as an optimization problem [157], which could minimize the frequency drop imposed by grid disturbance and the negative effect of SFD.

5.2 Centralized control

Different from the intrinsic response of distributed control, centralized control at the WF level first estimates system disturbance based on the measured frequency change and RoCoF. Then, the WF control center dispatches the command signals to the WTs determining the corresponding active power output during the frequency regulation [158]. References [159,160] introduced a sequential coordinated control strategy by adding a time delay, which prevented all WTs from entering the speed recovery stage at the same time. However, as the number of WTs increases in the WF, the WT speed recovery sequence arrangement will be more complicated, which limits its application in large-scale WF. To reduce the computation burden, a clustering-based optimization approach was proposed by first clustering the WTs into groups based on wind speed [161], where the same control command was dispatched to WTs in the same group. An improved system frequency response model was introduced in Ref. [162], which considered the inertia and droop control of WTs. To better estimate the virtual moment of inertia of WF, a matrix pencil method and the least squares algorithm were proposed [163]. A fuzzy logic gain control was proposed in Ref. [164] to mitigate the SFD with frequency deviation and RoCoF as input, and the sensitivity of the WT fatigue loads was considered during the optimization dispatch of active power [165,166]. References [167–169] combined the rotor kinetic energy control and de-loading control in the optimization problems of WF frequency regulation, with the wake effect taken into account, which not only mitigated the SFD, but also enhanced total energy harvesting of WF.

The abovementioned methods utilize one or more deflation approaches to simplify the optimization process, since the complex aerodynamic couplings and nonlinear characteristics of WTs during frequency regulation cannot be analytically modeled. Data-driven model can be an alternative way [170]. In Ref. [171], a distributed synchronized control method was introduced, which used phasor measurement unit (PMU) data in the WFs to calculate the optimum power share ratio. An active fault-tolerant cooperative control was proposed in Ref. [172] for large-scale WF clusters. A preview-based robust reinforcement learning method was proposed [173], in which, for the first time, a data-driven model-free solution was designed for the WF to participate in frequency regulation.

However, the data-driven model requires online-training to enhance the learning process and over-reliance on measured data brings the risk of cyber uncertainties in real-time control of power system [174]. Therefore, the model predictive control (MPC) algorithm has become a research hotspot in dealing with uncertainty modeling of system integration of large-scale WFs in recent years [175–181], which will be further discussed in Section 5.2.1.

5.2.1 Model predictive control in WF

Fig.17 summarizes the research framework of MPC-based active power control of WFs [175].

MPC adopts a rolling optimization strategy, replaces the global optimum with the local optimum, and uses the real-time measured information for feedback correction to enhance the robustness of the control [182,183]. Yin et al. [184] proposed a multi-objective predictive control strategy for WFs based on machine learning, and heuristic optimization was further established, thereby reducing computational costs and improving overall efficiency. References [185–187] proposed the integration of several WFs with similar locations which could be grouped as WP clusters. Based on the PMU as real-time data, a distributed MPC-based load frequency control model was established, which could keep the system frequency and inter-regional exchange power changing within a small range. Based on the ultra-short-term WP prediction information, the distributed MPC-based optimization mechanism of advanced frequency control was established in Ref. [188], and a Nash equilibrium decomposition coordinated online optimization control algorithm was used to solve the contradiction between the constrained multivariable complex system and the online rolling optimization solution. Guo et al. [189] proposed a double-layer MPC-based control strategy as shown in Fig.18. Compared with the conventional centralized method, the drawbacks due to forecast errors and data synchronization are largely offset.

To sum up, the distributed control can perform intrinsic frequency regulation without time delay, which can significantly arrest the RoCoF decline during inertia response, especially with the implementation of GFM control. The centralized control can better guarantee the power balance between the disturbance estimation and active power allocation in the WF, but the reliable communication requirement will increase the total investment cost. A distributed MPC-based hierarchical control scheme which combines the distributed control of GFM and the online power optimization of centralized control with sparse communication requirements will be the future trend.

5.3 WF with energy storage system

The strong randomness and fluctuation of WF power output pose great challenges to the frequency response of the power system, which brings application prospects for energy storage systems (ESSs) with fast response capabilities. Fig.19 shows the power rating range and power support time-scale of different types of ESSs in system frequency regulation [190–193].

Supercapacitor [194–196], battery [197–204], flywheel [205–211], compressed air [212–214], pump hydro [215,216], and superconductor magnetics energy storage (SMES) [217–219] have been used in a wide range of practices for WP frequency regulation. In Refs. [194,195], the fast response characteristics of supercapacitor were implemented for addressing the problem of SFD, where the supercapacitor unit was put in parallel on the DC-link bus of the AC-DC-AC power electronic converter of DFIG.

Due to its flexible capacity configuration and fast response speed, the battery energy storage system (BESS) has been widely utilized in wind-storage combined frequency regulation. However, the main drawback of BESS is cost. Moreover, the extensive use of charging and discharging will shorten its lifespan. To address this shortcoming, Refs. [202,203] proposed a state-machine-based coordinated control with an adaptive state of charge (SoC) feedback considering the real-time operating status of WTs and the SoC of BESS, thereby reducing the configuration size of the BESS and prolonging its service life.

Compared with BESS, flywheel can archive a fast response during PFR as well, but it is more durable and usually owns a longer lifespan [205,208]. Reference [210] mentioned that the energy stored in the flywheel in a short time could meet the lower frequency regulation demand, and the energy stored in the fuel cell for a long time could meet the higher frequency regulation demand. In addition, it proposed coordinated control with flywheel and fuel cell, which fully utilized the role of energy storage in a variety of frequency regulation requirements.

As a newly emerging form of energy storage, hydrogen not only provides clean energy for the future transportation industry, but also has a huge potential to participate in the system frequency regulation with WP [220,221]. References [222–224] proposed that the application of water electrolysis in a hydrogen filling station could behave as a demand-side response, and the simulation result showed that the system frequency stability was significantly improved under the scenario of 25% WP penetration. However, the alkaline electrolyzer requires a stable input power supply, which is contrary to the fluctuation characteristic of WP, and the proton exchange membrane and the solid oxide electrolyzer has high investment costs, which is not currently in large-scale commercial use [225,226].

In general, the ESS can effectively ease the randomness and fluctuation of WF power output and compensate for the drawbacks of WT-based frequency regulation, such as SFD. In addition, the hybrid ESS with different types of ESS, which can archive their complementary benefits in terms of response time, power rating range, lifespan, and ancillary services revenue, is worthy of further in-depth study.

6 Prospects, future challenges, and solutions

As previously discussed, the growing penetration of WP poses challenges to the frequency regulation and stability of the power system. This section elaborates on the current and future challenges of WP participation in system frequency regulation from the aspects of WT level, WF level, power system level, and revision of grid codes and standards, respectively.

6.1 Coordinated control at the WT level

Section 4 has explicitly introduced the principles and characteristics of different control strategies of WTs participating in frequency regulation. For de-loading control, future research should be conducted to determine the proper reserve margin which can balance the economic benefits and technical performance of WT. For the rotor kinetic energy control, its coordination with the reserve power provided by either de-loading control or ESSs can reduce the risk of SFD and fully exploit the rotor kinetic energy of WT during inertia response and PFR. Moreover, the emerging GFM has a remarkable stability in a low short circuit ratio system, and the reasonable share ratio of GFM and GFL in the system under different RES penetration levels should be further investigated to enhance the system frequency response capability in different application scenarios.

6.2 Application of data-driven artificial intelligence at the WF level

The nonlinear time-varying feature poses challenges to the modeling of WF. The implementation of advanced metering infrastructure, such as light detection and ranging sensors (LiDAR) aimed to reduce the wind speed measurement noise, can improve the modeling precision. In addition, the proposed data-driven model-free method based on historical data and ultra-short-term WP prediction can be an alternative approach. However, the current challenges are the contradiction between model accuracy and training time in solving optimization problems, and time delay and cyber security in communication. It is to be pointed out that with the development of artificial intelligence, its application can be a critical asset. For instance, the combination of feed-forward control, which needs to be developed ahead of time to sense the potential frequency disturbance based on wide-area monitoring system, and feedback correction based on local measurement can be a potential solution.

6.3 Coordination of WP at the power system level

As the penetration of WP increases, its interaction with other generation resources, such as thermal, hydro, flexible loads, is vital for the future, and the synergy of multiple resources can complement the advantages. Besides, the placement of WP shows strong spatial characteristics. For example, the abundant WP resources are located in the north, northeast, and northwest of China, which is thousands of miles away from the load centers on the south-east coast. Therefore, the constructions of large-scale WP clusters via HVDC transmission lines and offshore WP hubs are vigorously developed. However, power electronics-based HVDC links are prone to overvoltage and overcurrent issues that will result in permanent damage. Moreover, the growing harmonics and multi-frequency resonances introduced by power electronics in large-scale WFs may cause tripping of WTs and malfunction of protection systems, and future in-depth studies are necessary.

6.4 Revision of grid codes and standards

The existing grid codes and standards are drafted based on the scenario of traditional fossil fuel-based power sources as dominant. However, with the global energy structural transition toward more RES integrated with the power system, or even 100% RES in the near future, the existing grid codes and standards need to be continually updated to maintain suitability for the coming transition.

7 Conclusions

The transition from conventional energy toward clean and low carbon energy has resulted in the rapid development of the WP, leading to the high WP penetration level in the future power system. As a result, the capabilities of the inertia response and frequency regulation of the future power system will be deteriorated significantly due to the low inertia characteristic of the WP. For this reason, it is urgent to investigate and assess the capabilities of the WP participating in the inertia response and frequency regulation for the power system with high penetration level of WP generation.

This paper reviewed the fundamentals and applications of WP participating in the system frequency regulation from the WT level to the WF level, respectively. State-of-art control strategies were analyzed, and corresponding pros and cons were discussed and compared as well. In general, current research predominantly focus on the improvement of individual control strategies at the WT level only. However, from the perspective of power system frequency control, the overall contributions of frequency supporting at the WF level, which considers the dynamic aggregated control characteristics of WTs at the WF, are of major importance. Moreover, the complex aerodynamic interactions within the WF pose further challenges to the modeling precision and coordination of control strategies. It is expected that the data-driven model-free method can be an alternative approach, and the implementation of hybrid ESSs can compensate for the drawbacks of WP-based frequency control strategies.

The current and future challenges to be addressed still require further in-depth studies. For example, more attention ought to be paid to the real-time simulation, further hardware-in-loop testing platform, and actual application of demonstration project on the operation scenario of power system with higher penetration of WP. The impact of these on the system dynamics and frequency safety under non-ideal conditions needs to be assessed, and further ongoing technological enhancements and continuing research activities will promote the energy transition with large-scale interconnection of RES in future power systems.

References

[1] Dong J, Xue G, Dong M. . Energy-saving power generation dispatching in China: Regulations, pilot projects and policy recommendations—A review. Renewable & Sustainable Energy Reviews, 2015, 43: 1285–1300

[2] REN21. Renewables Global Status Report, REN21. 2022−8, available at REN21 website

[3] Shu Y, Zhang Z, Guo J. . Study on key factors and solution of renewable energy accommodation. Proceedings of the CSEE, 2017, 37(01): 1–9

[4] Shu Y, Xue Y, Cai B. . A review of energy transition analysis. Automation of Electric Power Systems, 2018, 42(10): 1–12

[5] NationalEnergy Administration. 2020 Wind Power Grid Operation. 2022−8, available at NEA website

[6] Departmentof IndustryScienceof AustraliaResources . Renewable energy target scheme. 2022, available at Australian government

[7] Chen Z, Guerrero J M, Blaabjerg F. A review of the state of the art of power electronics for wind turbines. IEEE Transactions on Power Electronics, 2009, 24(8): 1859–1875

[8] Zhang J, Li M. Analysis of the frequency characteristic of the power systems highly penetrated by new energy generation. Proceedings of the CSEE, 2020, 40(11): 3498–3507

[9] Xue Y, Lei X, Xue F. . A review on impacts of wind power uncertainties on power systems. Proceedings of the CSEE, 2014, 34(29): 5029–5040

[10] Sun H, Xu T, Guo Q. . Analysis on blackout in Great Britain power grid on August 9th, 2019 and its enlightenment to power grid in China. Proceedings of the CSEE, 2019, 39(21): 6183–6192

[11] Zeng H, Sun F, Li T. . Analysis of 9·28 blackout in south Australia and its enlightenment to China. Automation of Electric Power Systems, 2017, 41(13): 1–6

[12] AustralianEnergy Market Operator. AER Black System Event Compliance Report–—Investigation into the Pre-Event, System Restoration, and Market Suspension Aspects Surrounding the 28 September 2016 Event. Australia: Australian Energy Market Operator Ltd., 2016

[13] YangZ. Wind farm’s participating in grid frequency regulation through active power control. Dissertation for the Master’s Degree. Hangzhou: Zhejiang University, 2016 (in Chinese)

[14] EON Netz GmbH. Grid code-high and extra high voltage. 2006, available at tennettso website

[15] EskomSystem OperationsPlanningDivision. Grid code requirements for wind energy facilities connected to distribution or transmission system in south Africa. 2012, available at nersa website

[16] ScottishHydro-Electric Transmission Ltd. Scottish grid code. 2010, available at Scottish government

[17] Nordic grid code 2007 (Nordic collection of rules). 2022, available at tut website

[18] NationalGrid (Great Britain). Grid code documents: Connection conditions. 2009, available at the website of National Grid (Great Britain)

[19] Energinet. Grid connection of wind turbines to networks with voltage above 100 kV, Regulation TF 3.2.5. 2004, available at the website of energinet

[20] WangW SChi Y NLiY, . Technical Specification for Connecting Wind Farm to Power System—Part 1: On Shore Wind Power. Beijing: Standards Press of China, 2021 (in Chinese)

[21] TransEnergieH-Q. Technical requirements for the connection of generation facilities to the Hydro-Quebec transmission system: Supply requirements for wind generation. 2003, available at the website of hydroquebec

[22] Díaz-González F, Hau M, Sumper A. . Participation of wind power plants in system frequency control: Review of grid code requirements and control methods. Renewable & Sustainable Energy Reviews, 2014, 34: 551–564

[23] Tang W, Hu J, Chang Y. . Modeling of DFIG-based wind turbine for power system transient response analysis in rotor speed control timescale. IEEE Transactions on Power Systems, 2018, 33(6): 1–1

[24] Zhong Q, Weiss G. Synchronverters: inverters that mimic synchronous generators. IEEE Transactions on Industrial Electronics, 2011, 58(4): 1259–1267

[25] Wu Z, Gao W, Gao T. . State-of-the-art review on frequency response of wind power plants in power systems. Journal of Modern Power Systems and Clean Energy, 2018, 6: 1–16

[26] Fernández-Bustamante P, Barambones O, Calvo I. . Provision of frequency response from wind farms: A review. Energies, 2021, 14(20): 6689

[27] Ye Z, Xie Y, Zhu H. Review on power system frequency regulation with high wind power permeability. Energy and Power Engineering, 2018, 10(8): 366–382

[28] Ullah N R, Thiringer T, Karlsson D. Temporary primary frequency control support by variable speed wind turbines-potential and applications. IEEE Transactions on Power Systems, 2008, 23(2): 601–612

[29] KundurPBalu N JLaubyM G. Power System Stability and Control. New York: McGraw-hill, 1994

[30] MachowskiJBialek JBumbyJ. Power System Dynamics: Stability and Control. 2nd ed. John Wiley & Sons, 2011

[31] ZhangZ GChen G PLiM J, . aCode on Security and Stability for Power System. Beijing: Standards Press of China, 2019 (in Chinese)

[32] HatziargyriouN DMilanovicJ VRahmannC, . Stability definitions and characterization of dynamic behavior in systems with high penetration of power electronic interfaced technologies. Technical Report, Power System Dynamic Performance Committee (PSDP), 2020

[33] Sun H, Wang B, Li W. . Research on inertia system of frequency response for power system with high penetration electronics. Proceedings of the CSEE, 2020, 40(16): 5179–5192

[34] Gong Y, Zhou Q, Guo Q. . Improvement and application of frequency analysis modules for power system in high proportion of renewable energy situation. Power System Technology, 2021, 45(12): 4603–4612

[35] Wang B, Sun H, Li W. . Minimum inertia estimation of power system considering dynamic frequency constraints. Proceedings of the CSEE, 2022, 42(01): 114–127

[36] Ren K, Zhang D, Huang Y. . Large-scale system inertia estimation based on new energy output ratio. Power System Technology, 2022, 4: 1–9

[37] Zhang Z, Zhang N, Du E. . Review and countermeasures on frequency security issues of power systems with high shares of renewables and power electronics. Proceedings of the CSEE, 2022, 42(01): 1–25

[38] Ma N, Wang D. Extracting spatial-temporal characterstics of frequency dynamic in large-scale power grids. IEEE Transactions on Power Systems, 2019, 34(4): 2654–2662

[39] Badesa L, Teng F, Strbac G. Conditions for regional frequency stability in power system scheduling—Part I: Theory. IEEE Transactions on Power Systems, 2021, 36(6): 5558–5566

[40] Anderson P M, Mirheydar M. A low-order system frequency response model. IEEE Transactions on Power Systems, 1990, 5(3): 720–729

[41] Bolzoni A, Perini R. Feedback couplings evaluation on synthetic inertia provision for grid frequency support. IEEE Transactions on Energy Conversion, 2021, 36(2): 863–873

[42] Zhao D, Wang C, Xie J. . Transient stability assessment of auto encoder considering frequency shift of inertia center. Power System Technology, 2022, 46(2): 662–670

[43] Li W, Jin C, Wen K. . Active frequency response control under high-power loss. Automation of Electric Power Systems, 2018, 42(08): 22–30

[44] HansenL HHelle LBlaabjergF, . Conceptual Survey of Generators and Power Electronics for Wind Turbines. Technical Report, Risoe National Laboratory, Roskilde (Denmark), 2001

[45] BurnhamD JSantoso SMuljadiE. Variable rotor-resistance control of wind turbine generators. In: IEEE Power & Energy Society General Meeting, Calgary, AB, Canada, 2009

[46] Ma J, Zhao D, Yao L. . Analysis on application of a current-source based DFIG wind generator model. CSEE Journal of Power and Energy Systems, 2018, 4(3): 352–361

[47] Chinchilla M, Arnaltes S, Burgos J C. Control of permanent-magnet generators applied to variable-speed wind-energy systems connected to the grid. IEEE Transactions on Energy Conversion, 2006, 21(1): 130–135

[48] Ekanayake J, Jenkins N. Comparison of the response of doubly fed and fixed-speed induction generator wind turbines to changes in network frequency. IEEE Transactions on Energy Convention, 2004, 19(4): 800–802

[49] Xu L. Coordinated control of DFIG’s rotor and grid side converters during network unbalance. IEEE Transactions on Power Electronics, 2008, 23(3): 1041–1049

[50] Lima F K A, Luna A, Rodriguez P. . Rotor voltage dynamics in the doubly fed induction generator during grid faults. IEEE Transactions on Power Electronics, 2010, 25(1): 118–130

[51] Li S, Tang H, Deng C. . Calculation of wind power penetration limit involving frequency constraints and frequency regulation of wind turbines. Automation of Electric Power Systems, 2019, 43(4): 33–39

[52] Liu Z, Zhou M, Li Z. . Review of inertia control technology and requirement evaluation in renewable-dominant power system. Electric Power Automation Equipment, 2021, 41(12): 1–11

[53] Cardenas R, Pena R, Alepuz S. . Overview of control systems for the operation of DFIGs in wind energy applications. IEEE Transactions on Industrial Electronics, 2013, 60(7): 2776–2798

[54] Lu Z, Sheng W, Zhong Q. . Virtual synchronous generator and its applications in micro-grid. Proceedings of the CSEE, 2014, 34(16): 2591–2603

[55] HuJ. Investigation on the ride-through operation of DFIG-based wind power generation systems during grid fault—basic theory and key technology. Dissertation for the Docotoral Degree Hangzhou: Zhejiang University, 2009 (in Chinese)

[56] BianchiF DDe Battista HMantzR J. Wind Turbine Control Systems: Principles, Modeling and Gain Scheduling Design. Berlin: Springer Science & Business Media, 2006

[57] WuYLuG QuocD P. A review of frequency regulation of DFIG-based wind farms. In: The 11th IET International Conference on Advances in Power System Control, Operation and Management (APSCOM 2018), Hong Kong, China, 2018

[58] Liu J, Yao W, Wen J. . Prospect of technology for large-scale wind farm participating into power grid frequency regulation. Power System Technology, 2014, 38(3): 9

[59] Dreidy M, Mokhlis H, Mekhilef S. Inertia response and frequency control techniques for renewable energy sources: A review. Renewable & Sustainable Energy Reviews, 2017, 69: 144–155

[60] Zeng X, Liu T, Wang S. . Comprehensive coordinated control strategy of PMSG-based wind turbine for providing frequency regulation services. IEEE Access: Practical Innovations, Open Solutions, 2019, 7: 63944–63953

[61] Ochoa D, Martinez S. Fast-frequency response provided by DFIG-wind turbines and its impact on the grid. IEEE Transactions on Power Systems, 2017, 32(5): 4002–4011

[62] WenZYaoL LiangS, . Retrospect and prospect of wind power participating in power system frequency regulation. In: 2022 International Conference on Power Energy Systems and Applications (ICoPESA), Singapore, 2022

[63] Cheng Y, Azizipanah-Abarghooee R, Azizi S. . Smart frequency control in low inertia energy systems based on frequency response techniques: A review. Applied Energy, 2020, 279: 115798

[64] Thapa K B, Jayasawal K. Pitch control scheme for rapid active power control of a PMSG-based wind power plant. IEEE Transactions on Industry Applications, 2020, 56(6): 6756–6766

[65] TielensPRijcke S DSrivastavaK, . Frequency support by wind power plants in isolated grids with varying generation mix. In: Power & Energy Society General Meeting, San Diego, CA, USA, 2012

[66] Castro L M, Fuerte-Esquivel C R, Tovar-Hernandez J H. Solution of power flow with automatic load-frequency control devices including wind farms. IEEE Transactions on Power Systems, 2012, 27(4): 2186–2195

[67] Hu J, Xu G, Bi T. . A strategy of frequency control for deloaded wind turbine generator based on coordination between rotor speed and pitch angle. Power System Technology, 2019, 43(10): 3656–3663

[68] Lasheen A, Elshafei A L. Wind-turbine collective-pitch control via a fuzzy predictive algorithm. Renewable Energy, 2016, 87(MAR.PT.1): 298–306

[69] JanssensN ALambin GBragardN. Active power control strategies of DFIG wind turbines. In: 2007 IEEE Lausanne Power Technology, Lausanne, Switzerland, 2007

[70] Li S, Huang Y, Wang L. . Modeling primary frequency regulation auxiliary control system of doubly fed induction generator based on rotor speed control. Proceedings of the CSEE, 2017, 37(24): 7077–7086

[71] deAlmeida R G, Castronuovo E D, PecasLopes J A. Optimum generation control in wind parks when carrying out system operator requests. IEEE Transactions on Power Systems, 2006, 21(2): 718–725

[72] de Almeida R G, Pecas Lopes J A. Participation of doubly fed induction wind generators in system frequency regulation. IEEE Transactions on Power Systems, 2007, 22(3): 944–950

[73] Li S, Zhu G. Capability and effect of primary frequency regulation by wind turbine generators with active power reserve. Advanced Technology of Electrical Engineering and Energy, 2015, 34(10): 28–33, 50

[74] Mu G, Cai T, Yan G. . Bidirectional power constraints and influence of doubly fed induction generator participating in continuous frequency regulation. Transactions of China Electrotechnical Society, 2019, 34(08): 1750–1759

[75] WangM JShao Q F. Case analysis of wind turbine overspeed problem. Wind Energy (Chichester, England), 2019, (8): 5 (in Chinese)

[76] Mauricio J M, Marano A, Gomez-Exposito A. . Frequency regulation contribution through variable-speed wind energy conversion systems. IEEE Transactions on Power Systems, 2009, 24(1): 173–180

[77] Toulabi M, Ashouri-Zadeh A, Kazari H. . Application of Bang Bang controller to emulate primary frequency response in DFIGs. IEEE Systems Journal, 2020, 14(2): 1–9

[78] Ping K K, Pei L, Hadi B. . Kinetic energy of wind-turbine generators for system frequency support. IEEE Transactions on Power Systems, 2009, 24(1): 279–287

[79] ItaniS EAnnakkage U DJoosG. Short-term frequency support utilizing inertial response of DFIG wind turbines. In: Power and Energy Society General Meeting, Detroit, MI, USA, 2011

[80] Hansen A D, Altin M, Margaris I D. . Analysis of the short-term overproduction capability of variable speed wind turbines. Renewable Energy, 2014, 68: 326–336

[81] Hafiz F, Abdennour A. Optimal use of kinetic energy for the inertial support from variable speed wind turbines. Renewable Energy, 2015, 80: 629–643

[82] Teninge A, Jecu C, Roye D. . Contribution to frequency control through wind turbine inertial energy storage. IET Renewable Power Generation, 2010, 3(3): 358–370

[83] Margaris I D, Papathanassiou S A, Hatziargyriou N D. . Frequency control in autonomous power systems with high wind power penetration. IEEE Transactions on Sustainable Energy, 2012, 3(2): 189–199

[84] Vidyanandan K V, Senroy N. Primary frequency regulation by deloaded wind turbines using variable droop. IEEE Transactions on Power Systems, 2013, 28(2): 837–846

[85] Pan W, Quan R, Wang F. A variable droop control strategy for doubly-fed induction generators. Automation of Electric Power Systems, 2015, 39(11): 126–131, 186

[86] Van de Vyver J, De Kooning J D M, Meersman B. . Droop control as an alternative inertial response strategy for the synthetic inertia on wind turbines. IEEE Transactions on Power Systems, 2016, 31(2): 1129–1138

[87] Ashabani S M, Mohamed Y A R I. General interface for power management of micro-grids using nonlinear cooperative droop control. IEEE Transactions on Power Systems, 2013, 28(3): 2929–2941

[88] Mohamed A R I, El-Saadany E F. Adaptive decentralized droop controller to preserve power sharing stability of paralleled inverters in distributed generation microgrids. IEEE Transactions on Power Electronics, 2008, 23(6): 2806–2816

[89] Huang L, Xin H, Zhang L. . Synchronization and frequency regulation of DFIG-based wind turbine generators with synchronized control. IEEE Transactions on Energy Conversion, 2017, 32(3): 1251–1262

[90] Arani M F M, Mohamed Y A I. Dynamic droop control for wind turbines participating in primary frequency regulation in microgrids. IEEE Transactions on Smart Grid, 2018, 6: 5742–5751

[91] HurtadoSGostales Gde LaraA, . A new power stabilization control system based on making use of mechanical inertia of a variable speed wind turbine for stand-alone wind diesel applications. In: IEEE 28th Annual Conference of the Industrial Electronics Society, Sevilla, Spain, 2022

[92] Morren J, de Haan S W H, Kling W L. . Wind turbines emulating inertia and supporting primary frequency control. IEEE Transactions on Power Systems, 2006, 21(1): 433–434

[93] Li S, Wang W, Zhang X. . Impact of wind power on power system frequency and combined virtual inertia control. Automation of Electric Power Systems, 2019, 43(15): 64–70

[94] Tang X, Miao F, Qi Z. . Survey on frequency control of wind power. Proceedings of the CSSE, 2014, 34(25): 4304–4314

[95] Lan F, Pan Y, Shi M. . Optimal variable-coefficient virtual inertia control for DFIG-based wind turbines. Automation of Electric Power Systems, 2019, 43(12): 51–59

[96] Tian X, Wang W, Chi Y. . Adaptation virtual inertia control strategy of DFIG and assessment of equivalent virtual inertia time constant of connected power system. Journal of Engineering (Stevenage, England), 2017, 2017(13): 922–928

[97] Ding L, Yin S, Wang T. . Active rotor speed protection strategy for DFIG-based wind turbines with inertia control. Automation of Electric Power Systems, 2015, 39(24): 29–34, 95

[98] Qiao Y, Guo X, Lu Z. . Parameter setting of auxiliary frequency regulation of wind turbines considering secondary frequency drop. Power System Technology, 2020, 03: 807–815

[99] Liu B, Yang J, Liao K. . Improved frequency control strategy for DFIG-based wind turbines based on rotor kinetic energy control. Automation of Electric Power Systems, 2016, 40(16): 17–22

[100] Hou Y H, Sun L, Liu K. . Analysis and mitigation of electromechanical oscillations for DFIG wind turbines involved in fast frequency response. IEEE Transactions on Power Systems, 2020, 34(6): 4547–4556

[101] Zhang X, He W, Hu J. Impact of inertia control of DFIG-based WT on torsional vibration in drivetrain. IEEE Transactions on Sustainable Energy, 2020, 11(4): 2525–2534

[102] ZhangXHu JHeW. A damping control strategy for torsional oscillations of DFIG-based WTs caused by inertia control. In: 2020 IEEE Power & Energy Society General Meeting (PESGM), Montreal, Canada, 2020

[103] TayyebiADörfler FKupzogF, . Grid-forming converters—Inevitability, control strategies and challenges in future grids application. CIRED Workshop: paper 0236, 2018

[104] Zhang H, Xiang W, Lin W. . Grid forming converters in renewable energy sources dominated power grid: Control strategy, stability, application, and challenges. Journal of Modern Power Systems and Clean Energy, 2021, 9(6): 18

[105] Rosso R, Wang X, Liserre M. . Grid-forming converters: Control approaches, grid-synchronization, and future trends—A review. IEEE Open Journal of Industry Applications, 2021, 2: 93–109

[106] Lasseter R H, Chen Z, Pattabiraman D. Grid-forming inverters: A critical asset for the power grid. IEEE Journal of Emerging and Selected Topics in Power Electronics, 2020, 8(2): 925–935

[107] Tayyebi A, Gross D, Anta A. . Frequency stability of synchronous machines and grid-forming power converters. IEEE Journal of Emerging and Selected Topics in Power Electronics, 2020, 8(2): 1004–1018

[108] Pan D, Wang X, Liu F. . Transient stability of voltage-source converters with grid-forming control: A design-oriented study. IEEE Journal of Emerging and Selected Topics in Power Electronics, 2020, 8(2): 1019–1033

[109] De Brabandere K, Bolsens B, Van den Keybus J. . A voltage and frequency droop control method for parallel inverters. IEEE Transactions on Power Electronics, 2007, 22(4): 1107–1115

[110] Zhang L, Harnefors L, Nee H P. Power-synchronization control of grid-connected voltage-source converters. IEEE Transactions on Power Systems, 2010, 25(2): 809–820

[111] Arghir C, Dorfler F. The electronic realization of synchronous machines: Model matching, angle tracking, and energy shaping techniques. IEEE Transactions on Power Electronics, 2020, 35(4): 4398–4410

[112] Johnson B B, Dhople S V, Hamadeh A O. . Synchronization of parallel single-phase inverters with virtual oscillator control. IEEE Transactions on Power Electronics, 2014, 29(11): 6124–6138

[113] Ali M, Nurdin H I, Fletcher J E. Dispatchable virtual oscillator control for single-phase islanded inverters: Analysis and experiments. IEEE Transactions on Industrial Electronics, 2021, 68(6): 4812–4826

[114] Xi J, Geng H, Ma S. . Inertial response characteristics analysis and optimisation of PMSG-based VSG-controlled WECS. IET Renewable Power Generation, 2018, 12(15): 1741–1747

[115] Yan W, Cheng L, Yan S. . Enabling and evaluation of inertial control for PMSG-WTG using synchronverter with multiple virtual rotating masses in microgrid. IEEE Transactions on Sustainable Energy, 2020, 11(2): 1078–1088

[116] Jiao Y, Nian H. Grid-forming control for DFIG based wind farms to enhance the stability of LCC-HVDC. IEEE Access: Practical Innovations, Open Solutions, 2020, 8: 156752–156762

[117] WangSHu JYuanX. DFIG-based wind turbines with virtual synchronous control: Inertia support in weak grid. In: 2015 IEEE Power & Energy Society General Meeting, Denver, CO, USA, 2015

[118] Hu J, Sun L, Yuan X. . Modeling of type 3 wind turbines with df/dt inertia control for system frequency response study. IEEE Transactions on Power Systems, 2017, 32(4): 2799–2808

[119] LeiYHeW LiaoM, . Inertia characteristics analysis of DFIG-based wind turbines with virtual synchronous control in different operation areas. In: 2019 IEEE Sustainable Power and Energy Conference (iSPEC), Beijing, China, 2019

[120] Yu H, Awal M A, Tu H. . Comparative transient stability assessment of droop and dispatchable virtual oscillator controlled grid-connected inverters. IEEE Transactions on Power Electronics, 2021, 36(2): 2119–2130

[121] Xin H, Huang L, Zhang L. . Synchronous instability mechanism of P−f droop-controlled voltage source converter caused by current saturation. IEEE Transactions on Power Systems, 2016, 31(6): 5206–5207

[122] Wu H, Wang X. Design-oriented transient stability analysis of grid-connected converters with power synchronization control. IEEE Transactions on Industrial Electronics, 2019, 66(8): 6473–6482

[123] Paolone M, Gaunt T, Guillaud X. . Fundamentals of power systems modelling in the presence of converter-interfaced generation. Electric Power Systems Research, 2020, 189: 106811

[124] Zhang Z, Sun Y, Li G. . Frequency regulation by doubly fed induction generator wind turbines based on coordinated overspeed control and pitch control. Automation of Electric Power Systems, 2011, 35(17): 7

[125] Ke X, Zhang W, Li P. . Fuzzy adaptive virtual inertia control for high wind power penetration system. Power System Technology, 2020, 44(6): 8

[126] Li S, Wang W, Zhang X. . Fuzzy adaptive virtual inertia control strategy of wind turbines based on system frequency response interval division. Power System Technology, 2021, 45(5): 7

[127] Ashouri-Zadeh A, Toulabi M, Ranjbar A M. Coordinated design of fuzzy-based speed controller and auxiliary controllers in a variable speed wind turbine to enhance frequency control. IET Renewable Power Generation, 2016, 10(9): 1298–1308

[128] Ma S, Geng H, Yang G. . Virtual inertia control of variable-speed wind turbines based on the switching of power point tracking curves. Automation of Electric Power Systems, 2018, 42(17): 42–48

[129] Ghosh S, Kamalasadan S, Senroy N. . Doubly fed induction generator (DFIG)-based wind farm control framework for primary frequency and inertial response application. IEEE Transactions on Power Systems, 2016, 31(3): 1861–1871

[130] Fu Y, Wang Y, Zhang X. Integrated wind turbine controller with virtual inertia and primary frequency responses for grid dynamic frequency support. IET Renewable Power Generation, 2017, 11(8): 1129–1137

[131] Ding L, Yin S, Wang T. . Integrated frequency control strategy of DFIGs based on virtual inertia and over-speed control. Power System Technology, 2015, 39(9): 2385–2391

[132] Boyle J, Littler T, Muyeen S M. . An alternative frequency-droop scheme for wind turbines that provide primary frequency regulation via rotor speed control. International Journal of Electrical Power & Energy Systems, 2021, 133(1): 107219

[133] Ouyang J, Pang M, Li M. . Frequency control method based on the dynamic deloading of DFIGs for power systems with high-proportion wind energy. International Journal of Electrical Power & Energy Systems, 2021, 128(4): 106764

[134] Huang L, Xin H, Wang Z. . Transient stability analysis and control design of droop-controlled voltage source converters considering current limitation. IEEE Transactions on Smart Grid, 2019, 10(1): 578–591

[135] Verma P, K S, Dwivedi B. A cooperative approach of frequency regulation through virtual inertia control and enhancement of low voltage ride-through in DFIG-based wind farm. Journal of Modern Power Systems and Clean Energy, 2022, 10(6): 1519–1530

[136] Poolla B K, Gros D, Dorfler F. Placement and implementation of grid-forming and grid-following virtual inertia and fast frequency response. IEEE Transactions on Power Systems, 2019, 34(4): 3035–3046

[137] Qin S, Qin C, Li S. . Review of the voltage-source grid forming wind turbine. Proceedings of the CSEE, 2023, 43(04): 1314–1334

[138] Chen Z, Yin M, Zou Y. . Maximum wind energy extraction for variable speed wind turbines with slow dynamic behavior. IEEE Transactions on Power Systems, 2017, 32(4): 3321–3322

[139] Wang L, Liu J, Qian F. Wind speed frequency distribution modeling and wind energy resource assessment based on polynomial regression model. International Journal of Electrical Power & Energy Systems, 2021, 130(1): 106964

[140] Frehlich R, Kelley N. Measurements of wind and turbulence profiles with scanning doppler lidar for wind energy applications. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2008, 1(1): 42–47

[141] Mohammadi E, Fadaeinedjad R, Naji H R. . Investigation of horizontal and vertical wind shear effects using a wind turbine emulator. IEEE Transactions on Sustainable Energy, 2019, 10(3): 1206–1216

[142] Fadaeinedjad R, Moschopoulos G, Moallem M. The impact of tower shadow, yaw error, and wind shears on power quality in a wind–diesel system. IEEE Transactions on Energy Conversion, 2009, 24(1): 102–111

[143] Xie Z, Xu Z, Zhang X. . Improved power pulsation suppression of DFIG for wind shear and tower shadow effects. IEEE Transactions on Industrial Electronics, 2017, 64(5): 3672–3683

[144] Dolan D S L, Lehn P W. Simulation model of wind turbine 3p torque oscillations due to wind shear and tower shadow. IEEE Transactions on Energy Conversion, 2006, 21(3): 717–724

[145] Kim H, Singh C, Sprintson A. Simulation and estimation of reliability in a wind farm considering the wake effect. IEEE Transactions on Sustainable Energy, 2012, 3(2): 274–282

[146] Howlader A M, Senjyu T, Saber A Y. An integrated power smoothing control for a grid-interactive wind farm considering wake effects. IEEE Systems Journal, 2015, 9(3): 954–965

[147] Wang Z, Wu W. Coordinated control method for DFIG-based wind farm to provide primary frequency regulation service. IEEE Transactions on Power Systems, 2018, 33(3): 2644–2659

[148] Zhang X, Chen Y, Wang Y. . Deloading power coordinated distribution method for frequency regulation by wind farms considering wind speed differences. IEEE Access: Practical Innovations, Open Solutions, 2019, 7: 122573–122582

[149] Chang-Chien L R, Yin Y C. Strategies for operating wind power in a similar manner of conventional power plant. IEEE Transactions on Energy Conversion, 2009, 24(4): 926–934

[150] Lee J, Muljadi E, Sorensen P. . Releasable kinetic energy-based inertial control of a DFIG wind power plant. IEEE Transactions on Sustainable Energy, 2016, 7(1): 279–288

[151] Yang P, Dong X, Li Y. . Research on primary frequency regulation control strategy of wind-thermal power coordination. IEEE Access: Practical Innovations, Open Solutions, 2019, 7: 144766–144776

[152] Xue L, Jia Y, Dong Z. Adaptive frequency responsive control for wind farm considering wake interaction. Journal of Modern Power Systems and Clean Energy, 2021, 9(5): 1066–1075

[153] Wu Y, Zeng J, Lu G. . Development of an equivalent wind farm model for frequency regulation. IEEE Transactions on Industry Applications, 2020, 56(3): 2360–2374

[154] Shi Q, Wang G, Li H. . Coordinated virtual inertia control strategy of multiple wind turbines in wind farms considering frequency regulation capability. Power System Technology, 2019, 43(11): 4005–4017

[155] Wu Y, Yang W, Hu Y. . Frequency regulation at a wind farm using time-varying inertia and droop controls. IEEE Transactions on Industry Applications, 2019, 55(1): 213–224

[156] Liu H, Liu X, Xiong L. . Adaptive power compensation-based frequency regulation strategy of wind turbine system. IEEE Journal on Emerging and Selected Topics in Circuits and Systems, 2022, 12(1): 260–267

[157] Kheshti M, Ding L, Bao W. . Toward intelligent inertial frequency participation of wind farms for the grid frequency control. IEEE Transactions on Industrial Informatics, 2020, 16(11): 6772–6786

[158] Baccino F, Conte F, Grillo S. . An optimal model-based control technique to improve wind farm participation to frequency regulation. IEEE Transactions on Sustainable Energy, 2015, 6(3): 993–1003

[159] He C, Wang H, Sun H. . Analysis on frequency control characteristics of variable speed wind turbines and coordinated frequency control strategy of wind farm. Automation of Electric Power Systems, 2013, 37(09): 1–6, 59

[160] Zhang J, Li W, Wang H. . Multi-source cascaded frequency modulation scheme and time-sequence optimization strategy of frequency modulation at level of wind farm. Automation of Electric Power Systems, 2019, 43(15): 93–100

[161] Ma S, Geng H, Yang G. . Clustering-based coordinated control of large-scale wind farm for power system frequency support. IEEE Transactions on Sustainable Energy, 2018, 9(4): 1555–1564

[162] Ye H, Pei W, Qi Z. Analytical modeling of inertial and droop responses from a wind farm for short-term frequency regulation in power systems. IEEE Transactions on Power Systems, 2016, 31(5): 3414–3423

[163] Chen P, Qi C, Chen X. Virtual inertia estimation method of DFIG-based wind farm with additional frequency control. Journal of Modern Power Systems and Clean Energy, 2021, 9(5): 1076–1087

[164] Badihi H, Zhang Y, Hong H. Active power control design for supporting grid frequency regulation in wind farms. Annual Reviews in Control, 2015, 40: 70–81

[165] Liu Y, Wang Y, Wang X. . Active power dispatch for supporting grid frequency regulation in wind farms considering fatigue load. Energies, 2019, 12(8): 1508

[166] Wang X, Wang Y, Liu Y. Dynamic load frequency control for high-penetration wind power considering wind turbine fatigue load. International Journal of Electrical Power & Energy Systems, 2020, 117: 105696

[167] Ahmadyar A, Verbič G. Coordinated operation strategy of wind farms for frequency control by exploring wake interaction. IEEE Transactions on Sustainable Energy, 2017, 8(1): 230–238

[168] Liu T, Pan W, Quan R. . A variable droop frequency control strategy for wind farms that considers optimal rotor kinetic energy. IEEE Access: Practical Innovations, Open Solutions, 2019, 7: 68636–68645

[169] Lyu X, Jia Y, Xu Z. A novel control strategy for wind farm active power regulation considering wake interaction. IEEE Transactions on Sustainable Energy, 2020, 11(2): 618–628

[170] Jiao X, Geng H, Ma S. . Data-driven cooperative control method for multiple wind turbines. Journal of Power Supply, 2020, 18(2): 8

[171] Mahish P, Pradhan A K. Distributed synchronized control in grid integrated wind farms to improve primary frequency regulation. IEEE Transactions on Power Systems, 2020, 35(1): 362–373

[172] Jadidi S, Badihi H, Zhang Y. Fault-tolerant cooperative control of large-scale wind farms and wind farm clusters. Energies, 2021, 14(21): 1–29

[173] Dong H, Zhao X. Wind-farm power tracking via preview-based robust reinforcement learning. IEEE Transactions on Industrial Informatics, 2022, 18(3): 1706–1715

[174] Liu Y, Jiang H, Peng J. . A grouping-based frequency support scheme for wind farm under cyber uncertainty. IEEE Access: Practical Innovations, Open Solutions, 2020, 8: 178044–178054

[175] Ye L, Lu P, Dai B. . Review of model predictive control for power system with large-scale wind power grid-connected. Proceedings of the CSEE, 2021, 41(18): 6181–6198

[176] Kou P, Liang D, Yu L. . Nonlinear model predictive control of wind farm for system frequency support. IEEE Transactions on Power Systems, 2019, 34(5): 3547–3561

[177] Bao W, Wu Q, Ding L. . A hierarchical inertial control scheme for multiple wind farms with BESSs based on ADMM. IEEE Transactions on Sustainable Energy, 2021, 12(2): 751–760

[178] Soliman M, Malik O P, Westwick D T. Multiple model predictive control for wind turbines with doubly fed induction generators. IEEE Transactions on Sustainable Energy, 2011, 2(3): 215–225

[179] Li G, Ye H. Hierarchical nonlinear model predictive control for frequency support of wind farm. International Journal of Electrical Power & Energy Systems, 2021, 129(4): 106820

[180] Zhang B, Chen J, Wu W. A hierarchical model predictive control method for active power for accommodating large-scale wind power integration. Automation of Electric Power Systems, 2014, 38(09): 6–14

[181] Lu P, Ye L, Pei M. . Coordinated control strategy for active power of wind power cluster based on model predictive control. Proceedings of the CSEE, 2021, 41(17): 5887–5900

[182] Baccino F, Conte F, Grillo S. . An optimal model-based control technique to improve wind farm participation to frequency regulation. IEEE Transactions on Sustainable Energy, 2015, 6(3): 993–1003

[183] Ye L, Ren C, Li Z. . Stratified progressive predictive control strategy for multi-objective dispatching active power in wind farm. Proceedings of the CSEE, 2016, 36(23): 6327–6336, 6597

[184] Yin X, Zhao X, Lin J. . Reliability aware multi-objective predictive control for wind farm based on machine learning and heuristic optimizations. Energy, 2020, 202: 117739

[185] Zhang Y, Liu X, Qu B. . Distributed model predictive load frequency control of multi-area power system with DFIGs. IEEE/CAA Journal of Automatica Sinica, 2017, 4(1): 125–135

[186] Yang D, Cai G. Decentralized model predictive control based load frequency control for high wind power penetrated power systems. Proceedings of the CSEE, 2015, 35(03): 583–591

[187] Ma M, Liu X, Zhang C. LFC for multi-area interconnected power system concerning wind turbines based on DMPC. IET Generation, Transmission & Distribution, 2017, 11(10): 2689–2696

[188] Sun B, Tang Y, Zhong W. . Multi-area interconnected power system advanced frequency control strategy considering large scale wind power cluster integration based on DMPC. Proceedings of the CSEE, 2017, 37(21): 6291–6302

[189] Guo L, Ren Y, Wang Z. . Double-layer feedback control method for synchronized frequency regulation of PMSG-based wind farm. IEEE Transactions on Sustainable Energy, 2021, 12(4): 2423–2435

[190] Guan J, Gao C. An overview on the application of energy storage technologies to damp wind power output fluctuation. Power System and Clean Energy, 2011, 27(4): 6

[191] Wang H, Jiang Q. An overview of control and configuration of energy storage system used for wind power fluctuation mitigation. Automation of Electric Power Systems, 2014, 38(19): 126–135

[192] Ren G, Ma G, Cong N. Review of electrical energy storage system for vehicular applications. Renewable & Sustainable Energy Reviews, 2015, 41(C): 225–236

[193] Khodadoost Arani A A, Gharehpetian G B, Abedi M. Review on energy storage systems control methods in microgrids. International Journal of Electrical Power & Energy Systems, 2019, 107: 745–757

[194] Yan X, Sun W, Cui S. . Improved control strategy for inertia and primary frequency regulation of doubly fed induction generator based on rotor kinetic energy and supercapacitor energy storage. Transactions of China Electrotechnical Society, 2021, 36(S1): 179–190

[195] Zhou T, Sun W. Study on virtual inertia control for DFIG-based wind farms with high penetration. Proceedings of the CSEE, 2017, 37(02): 486–496

[196] Mohammadi E, Rasoulinezhad R, Moschopoulos G. Using a supercapacitor to mitigate battery microcycles due to wind shear and tower shadow effects in wind-diesel microgrids. IEEE Transactions on Smart Grid, 2020, 11(5): 3677–3689

[197] Sebastián R. Application of a battery energy storage for frequency regulation and peak shaving in a wind diesel power system. IET Generation, Transmission & Distribution, 2016, 10(3): 764–770

[198] Yu L, Kou P, Feng T. . Model predictive control strategy for combined wind-storage system to participate in frequency response. Automation of Electric Power Systems, 2019, 43(12): 36–43

[199] Yang G, Wang Y, Zhong C. . Frequency control strategy for wind storage combined system. Electric Power Construction, 2016, 37(12): 55–60

[200] Hu Z, Xia R, Wu L. . Joint operation optimization of wind-storage union with energy storage participating frequency regulation. Power System Technology, 2016, 40(08): 2251–2257

[201] Zhang C, Liu L, Cheng H. . Frequency-constrained co-planning of generation and energy storage with high-penetration renewable energy. Journal of Modern Power Systems and Clean Energy, 2021, 9(4): 760–775

[202] Liang C, Wang P, Han X. . Battery energy storage selection based on a novel intermittent wind speed model for improving power system dynamic reliability. IEEE Transactions on Smart Grid, 2018, 9(6): 6084–6094

[203] Tan J, Zhang Y. Coordinated control strategy of a battery energy storage system to support a wind power plant providing multi-timescale frequency ancillary services. IEEE Transactions on Sustainable Energy, 2017, 8(3): 1140–1153

[204] Xiong L, Yang S, He Y. . Specified time consensus control of ESSs for frequency support with DFIG wind turbines. International Journal of Electrical Power & Energy Systems, 2021, 135(2): 107520

[205] Ghosh S, Kamalasadan S. An integrated dynamic modeling and adaptive controller approach for flywheel augmented DFIG based wind system. IEEE Transactions on Power Systems, 2017, 32(3): 2161–2171

[206] Diaz-Gonzalez F, Bianchi F D, Sumper A. . Control of a flywheel energy storage system for power smoothing in wind power plants. IEEE Transactions on Energy Conversion, 2014, 29(1): 204–214

[207] Lai J, Song Y, Du X. Hierarchical coordinated control of flywheel energy storage matrix systems for wind farms. IEEE/ASME Transactions on Mechatronics, 2018, 23(1): 48–56

[208] Suvire G O, Molina M G, Mercado P E. Improving the integration of wind power generation into AC microgrids using flywheel energy storage. IEEE Transactions on Smart Grid, 2012, 3(4): 1945–1954

[209] Adhikari S, Karki R. Integrated disturbance response modeling of wind-integrated power systems to quantify the operational reliability benefits of flywheel energy storage. IEEE Transactions on Sustainable Energy, 2019, 10(3): 1152–1160

[210] Abdeltawab H H, Mohamed A. Robust energy management of a hybrid wind and flywheel energy storage system considering flywheel power losses minimization and grid-code constraints. IEEE Transactions on Industrial Electronics, 2016, 63(7): 4242–4254

[211] Islam F, Al-Durra A, Muyeen S M. Smoothing of wind farm output by prediction and supervisory-control-unit-based FESS. IEEE Transactions on Sustainable Energy, 2013, 4(4): 925–933

[212] Aldaadi M, Al-Ismail F, Al-Awami A T. . A coordinated bidding model for wind plant and compressed air energy storage systems in the energy and ancillary service markets using a distributionally robust optimization approach. IEEE Access: Practical Innovations, Open Solutions, 2021, 9: 148599–148610

[213] Calero I, Cañizares C A, Bhattacharya K. Compressed air energy storage system modeling for power system studies. IEEE Transactions on Power Systems, 2019, 34(5): 3359–3371

[214] Abouzeid S, Guo Y, Zhang H. Cooperative control framework of the wind turbine generators and the compressed air energy storage system for efficient frequency regulation support. International Journal of Electrical Power & Energy Systems, 2021, 130: 106844

[215] SalimathGSingh N KBadgeS S. Coordination and performance analysis of pumped hydro storage system integrated with solar, wind hybrid system. In: 2017 International Conference on Energy, Communication, Data Analytics and Soft Computing (ICECDS), Chennai, India, 2017

[216] BendibCKesraoui M. Wind-solar power system associated with flywheel and pumped-hydro energy storage. In: 2019 10th International Renewable Energy Congress (IREC), Sousse, Tunisia, 2019

[217] Yunus A M S, Masoum M A S, Abu-Siada A. Application of SMES to enhance the dynamic performance of DFIG during voltage sag and swell. IEEE Transactions on Applied Superconductivity, 2012, 22(4): 5702009

[218] Miao L, Wen J, Xie H. . Coordinated control strategy of wind turbine generator and energy storage equipment for frequency support. IEEE Transactions on Industry Applications, 2015, 51(4): 2732–2742

[219] Musarrat M N, Islam M R, Muttaqi K M. . Enhanced frequency support from a PMSG-based wind energy conversion system integrated with a high temperature SMES in standalone power supply systems. IEEE Transactions on Applied Superconductivity, 2019, 29(2): 1–6

[220] Pei W, Zhang X, Deng W. . Review of operational control strategy for DC microgrids with electric-hydrogen hybrid storage systems. CSEE Journal of Power and Energy Systems, 2022, 8(2): 329–346

[221] Dozein M G, Jalali A, Mancarella P. Fast frequency response from utility-scale hydrogen electrolyzers. IEEE Transactions on Sustainable Energy, 2021, 12(3): 1707–1717

[222] Kiaee M, Cruden A, Infield D. . Utilisation of alkaline electrolysers to improve power system frequency stability with a high penetration of wind power. IET Renewable Power Generation, 2014, 8(5): 529–536

[223] Fan F, Zhang R, Xu Y. . Robustly coordinated operation of an emission-free microgrid with hybrid hydrogen-battery energy storage. CSEE Journal of Power and Energy Systems, 2022, 8(2): 369–379

[224] VeerakumarNAhmad ZAdabiM E, . Fast active power-frequency support methods by large scale electrolyzers for multi-energy systems. In: 2020 IEEE PES Innovative Smart Grid Technologies Europe, Hague, Netherlands, 2020

[225] Guo X, Wei Y, Wang Y. . Review on power electronic converters for hydrogen production from renewable energy sources. Automation of Electric Power Systems, 2021, 45(20): 185–199

[226] Sun H, Li Z, Chen A. . Current status and development trend of hydrogen production technology by wind power. Transactions of China Electrotechnical Society, 2019, 34(19): 4071–4083

RIGHTS & PERMISSIONS

Higher Education Press 2023