Comparison of Dynamic Response Characteristics of Typical Energy Storage Technologies for Suppressing Wind Power Fluctuation

Abstract

The intermittence and randomness of wind speed leads to the fluctuation of wind turbine output power. In order to study the applicability of battery, super capacitor and flywheel energy storage technology in suppressing wind power fluctuation, this paper takes a 3 MW direct drive wind turbine as an example, and, through the establishment of a wind storage system model, the dynamic response characteristics and application effects of the three typical energy storage technologies to suppress the power fluctuation of the wind turbine under two wind speed fluctuation scenarios are simulated and studied, and the stability of output power is quantitatively analyzed. Results show that all the three energy storage systems respond well to power command curves, but when the wind power fluctuation is large, the flywheel energy storage has a better effect on suppressing the wind power fluctuation, which can suppress about 70% of the power fluctuation.

1. Introduction

As large-scale wind turbines are connected to the power grid, the fluctuation of their output has an increasingly significant impact on the stable operation and power quality of the power grid. In recent years, electrochemistry, super capacitor, flywheel and other energy storage technologies have been rapidly developed and applied. Because of its ability to absorb or release energy quickly for a short time, an energy storage system has become an effective means for suppressing wind power fluctuations and improving the acceptance capacity of power grid to wind power [1,2,3,4].

Taking the service life and economy of energy storage system as the objective, scholars have put forward some corresponding power suppression strategies and configuration modes considering the capacity and output of an energy storage system and a wind power system, respectively. In [5,6], multi-objective optimization methods of the energy storage system were proposed so as to alleviate voltage deviation and suppress power fluctuation. In [7], a kind of receding horizon control (RHC) scheme to seek a decision strategy to manage the operation of battery storage system was proposed, which can increase daily profits. An optimal power control strategy based on first-order low-pass filter was proposed in [8], which can minimize battery capacity by adjusting the filter smoothing time constant, thus achieving economical operation. In recent years, the flywheel energy storage system has been studied to suppress wind power fluctuations and improve the power quality of the power grid, but the present research is still in the theoretical or demonstration stage [9,10]. In [11,12], a coordinated control strategy of a hybrid energy storage system was proposed to smooth the output fluctuation of wind farms, which was based on the advantages and characteristics of various energy storage technologies. In [13], fluctuation degree was used to classify wind power fluctuation scenarios, to schedule energy storage output and to improve wind power schedulability. There are many studies on the economy and the service life of energy storage systems in the joint operation of an energy storage system and a wind energy system. By establishing economic models and considering the respective capacity and output of wind power systems and energy storage systems, the corresponding control strategies were proposed. In [14,15,16], the operation and planning models of energy storage systems and the shortcomings of existing models were summarized. As for the research on the response characteristics of energy storage systems to suppress the output fluctuation of new energy, the energy storage response time of MW-level BESS (Battery Storage System) in a photovoltaic-storage power station under different power switching was analyzed and compared in [17]. The power response characteristics of BESS were studied based on the operating characteristics of battery and the power regulation range of the converter in [18]. Taking a 50 MW wind farm in China as an example, different schemes such as single-battery energy storage, single-supercapacitor energy storage and battery–supercapacitor hybrid energy storage were compared in [19]. Under the same stabilization results, the annual equivalent cost of hybrid energy storage is significantly lower than that of single energy storage. At present, most of the studies on wind power accommodation by energy storage systems remain at the level of optimal scheduling and lack the refined modeling of energy storage systems, which cannot reflect the real-time voltage and power fluctuation information of the energy storage system [20]. Besides, the studies on the effects and comparison of various energy storage systems for the suppression of wind power fluctuations under different scenarios are not detailed.

The dynamic response characteristics of the energy storage system are important factors affecting the power suppression effect. Due to the differences in the characteristics of various energy storage technologies [21], the power suppression effects are slightly different in different scenarios. Therefore, in order to study the applicability of battery, super capacitor and flywheel energy storage technology in suppressing wind power fluctuation, this paper establishes a wind energy storage system model in PSCAD software. Through simulation, the dynamic response characteristics and effects of the three energy storage systems to suppress the output power fluctuation of the wind turbine in two wind speed fluctuation scenarios were analyzed and compared. Moreover, the stability of output power was quantitatively analyzed, which provided a certain reference for the selection of the energy storage configuration mode of the wind farm.

2. Mathematical Model and Control Strategy of Permanent Magnet Direct Drive Wind Turbine

Compared with the doubly-fed wind turbine, the permanent magnet direct drive wind turbine has a simple structure without brushes and gearboxes, low power transmission loss and low maintenance cost. Therefore, the permanent magnet direct drive wind turbine is selected as the wind turbine for the simulation in this paper; its structure diagram is shown in Figure 1.

2.1. Aerodynamic Model

The actual absorbed power of the wind turbine is shown as follows [22]:

$$P_m=C_p{P}_w=\frac{1}{2}\rho A{v}^3{C}_p$$

(1)

where ρ is air density, kg/m³; A is the area swept by the wind turbine wheel, m²; v is the upstream wind speed, m/s; C_p is wind energy utilization coefficient, which is a function related to tip speed ratio and pitch angle.

The formula of C_p is shown as follows [22]:

$$\{\begin{array}{l}{C}_p=0.22(\frac{116}{{\lambda }_i}-0.4\beta -5)e^{\frac{-12.5}{{\lambda }_i}}\\ \frac{1}{{\lambda }_i}=\frac{1}{\lambda +0.08\beta }-\frac{0.035}{{\rho }^3+1}\end{array}$$

(2)

where β is the pitch angle; ω is the rotational angular velocity of the wind wheel, rad/s; R is the radius of the wind wheel, m. v is the upwind wind speed, m/s. λ is the tip speed ratio.

The formula of λ is shown as follows [22]:

$$\lambda =\frac{\omega \cdot R}{\upsilon }$$

(3)

where R is the radius of the blade, m; ω is the mechanical angular velocity of the wind turbine, rad/s.

From the Equations (1)–(3), it can be concluded that the wind energy utilization rate of the wind turbine is related to the tip speed ratio when the wind speed is constant under the condition of fixed pitch angle. When the wind speed changes, the rotor speed is adjusted by constructing the PMSG controller to make the wind turbine work at the state of optimal tip speed ratio, so as to realize the maximum wind energy tracking.

2.2. Model of Permanent Magnet Synchronous Wind Generator System

In order to simplify the analysis, PMSG is assumed to be an ideal motor, the mathematical model of PMSG is established in the d-q synchronous rotation coordinate system, its voltage equation is shown as follows [23]:

$$\{\begin{array}{l}{u}_d=i_d{R}_s+L_d\frac{d{i}_d}{dt}-{\omega }_e{L}_q{i}_q\\ u_q=i_q{R}_s+L_q\frac{d{i}_q}{dt}+{\omega }_e(L_d{i}_d+{\psi }_f)\end{array}$$

(4)

where u_d and u_q are voltage components of d-q axis, respectively; R_s is stator resistance; i_d and i_q are current components of d-q axis, respectively; ω_e is electric angular velocity; L_d and L_q are inductance components of d-q axis, respectively and ψ_f is permanent magnet flux.

Due to the large number of pole pairs of the permanent magnet synchronous generator, the rotor permanent magnet generally adopts the patch type, which can ignore the salient pole effect. If x_d = x_q, the PMSG torque equation is shown as follows [23]:

$$T_e=\frac{3}{2}{n}_p{i}_q{\psi }_f$$

(5)

where n_p is the polar pairs of the motor.

2.3. Control Strategy of Converter

The bidirectional back-to-back converter is selected as the converter part of the direct drive wind turbine. Since the back-to-back full-power converter isolates the generator from the grid, if the influence of small fluctuation of DC capacitor voltage is ignored, the grid-connected electrical characteristics of the direct-drive wind turbine are mainly determined by the grid-side converter control system. Therefore, the generator-side converter control system can be appropriately simplified and the active and reactive power control of the wind turbine can be simulated by the grid-side converter.

The most common control strategy of grid-side converter is the vector control strategy of grid voltage orientation based on coordinate transformation theory. The voltage orientation control usually adopts the structure of the outer voltage loop and inner current loop. The control model of grid-side converter is shown in Figure 2. u_gd, u_gq are the d-axis and q-axis components of the grid voltage; i_gd, i_gq are the components of the output current of grid-side converter on the d-axis and q-axis; i_gdref, i_gqref are the reference values of output current of d-axis and q-axis grid-side converter; L_g is the equivalent inductance of the grid-side converter filter; w_s is the grid frequency; u_dc, u_dcref are the actual and reference values of the DC capacitance current; v_cd, v_cq are the d-axis and q-axis components of the AC port voltage of the grid-side converter. The output of the voltage outer loop is i_gdref, which is the given value of the active current; the reactive current i_gqref is set to 0 in the unit power factor control mode. After the active current and reactive current are controlled by the current inner loop feedback, the closed-loop output control variables v_cd and v_cq are connected to the PWM strategy interface to obtain the corresponding switching function.

3. Theoretical Analysis and Control Strategy of Energy Storage System

Compared with other types of energy storage technologies, super capacitor, battery and flywheel energy storage systems have faster response speeds, so they can timely respond to fast fluctuating power disturbance. Therefore, this paper selects the above three energy storage technologies to study their dynamic response characteristics and effects on suppressing the output power fluctuation of the wind turbine. The equivalent circuit models and corresponding control strategies of the three energy storage technologies are as follows.

3.1. Equivalent Circuit Model of Battery

There are many kinds of battery energy storage. At present, lead-acid storage battery technology tends to be mature and the cost is low, which has been applied in engineering practices on a large scale. This paper mainly considers the external power characteristics of the battery, so the improved basic model is selected as the equivalent circuit model of battery. The equivalent circuit model of lead–acid storage battery is shown in Figure 3.

There is a certain numerical relationship between the open circuit voltage E of the battery and its real-time SOC, the formula of E is shown as follows [24]:

$$\mathrm{E}={\mathrm{E}}_0-K\frac{Q_{bat}}{Q_{bat}-{\displaystyle \int idt}}+A\exp (-B{\displaystyle \int idt})$$

(6)

where E₀ is the internal electromotive force of the battery; K is the internal polarization voltage of the battery; Q_bat is rated capacity of the battery; i is the current in the equivalent circuit model of the battery; A is the voltage sag in the exponential region and B is the reciprocal of the capacity of the exponential region, (Ah)⁻¹.

3.2. Equivalent Circuit Model of Super Capacitor

The simulation in this paper mainly considers the external power characteristics of the supercapacitor, so the lumped parameter model is selected as the equivalent circuit model of the supercapacitor, as shown in Figure 4. In addition to the supercapacitor body, there are two resistors in the circuit model, which respectively represent the output internal resistance and static loss.

3.3. Selection of Flywheel Energy Storage Motor

The flywheel energy storage system is a mechanical rotating energy storage device. Based on the acceleration or deceleration rotation of the flywheel in different scenarios, it can complete rapid electromechanical energy storage and conversion. The motor of the flywheel energy storage system is required to run in the states of power generation and electric. In order to meet the requirements of large speed range, small no-load loss, high operation efficiency and long life, a permanent magnet synchronous motor is selected as the motor of flywheel energy storage system. At the same time, the flywheel body is simulated by increasing the inertia of permanent magnet synchronous motor. Through the corresponding control strategy, the flywheel switches between the charging and discharging states. The working state of the motor is shown in Figure 5.

3.4. Control Strategy of Energy Storage Power Conversion System

The power conversion system composed of power electronic devices is a key component of energy storage system, which undertakes the functions of AC and DC form conversion and bidirectional energy transfer between energy storage devices and power grid. When the energy storage system is battery or supercapacitor energy storage, the power conversion circuit structure is the “DC boost- inverter” circuit, and its structure is shown in Figure 6a. When the energy storage system is flywheel energy storage, the power conversion circuit structure is “rectifier-DC boost-inverter” circuit, the structure is shown in Figure 6b.

• Bidirectional DC/DC converter control

When the energy storage system is charged, the bidirectional DC/DC converter is used as buck circuit. When the energy storage system discharges, the bidirectional DC/DC converter is used as boost circuit. The main control objective of the bidirectional DC/DC converter is to maintain the DC side voltage constant, thereby reducing the voltage and frequency fluctuations of the whole system. The bidirectional DC/DC converter adopts the control mode of voltage outer loop and current inner loop and the control schematic diagram is shown in Figure 7. When the output voltage of DC side is less than the reference voltage, the difference value between actual voltage and reference voltage is input into the PI controller to obtain the current reference value of the battery or supercapacitor side. Then, the difference value between actual current and reference current is input into the PI controller, and the value of the duty cycle is calculated by the set parameters. After the comparison with the triangular carrier, the PWM signal is output to control the turn-on and turn-off of S1. At this time, the DC/DC converter is used as the boost circuit. When the output voltage of the DC side is larger than the reference voltage, the control circuit controls the turn-on and turn-off of S2, which is the same as the above principle. At this time, the DC/DC converter is used as the buck circuit.

2.

Hysteresis current control

The control adopted in the inverter part is hysteresis comparison, that is, the actual current value is compared with the preset current curve. When the actual value is larger than the set value, the upper half bridge of IGBT is turned off and the lower half bridge of IGBT is turned on. When the actual value is less than the set value, the upper half bridge of IGBT is turned on and the lower half bridge of IGBT is turned off. Through the control, the current waveform is consistent with the set waveform. The set current waveform is calculated from the power, so the energy storage system can output specific power by controlling the current and achieving the purpose of suppressing the wind power fluctuation. The difference between the reference output power of PCC and the actual output power of wind turbine is used as the reference output power of the energy storage system in the simulation. Hysteresis current control structure diagram is shown in Figure 8.

4. Case Study

4.1. Simulation of Wind Storage System

The wind storage system model is established in PSCAD software and its overall structure is shown in Figure 9. The rated power of wind turbine is 3 MW and the main parameter settings of energy storage system are shown in

Table 1. Main parameters of energy storage system.

Parameter Name	Value	Parameter Name	Value
Battery rated voltage	400 V	Supercapacitor capacity	3000 F
Battery internal resistance	0.1 Ω	Supercapacitor internal resistance	0.45 mΩ
Battery rated power	2 MW	Supercapacitor rated voltage	2 V
Battery rated capacity	4 MWh	Supercapacitor bank	200 (series) × 20 (parallel)
Flywheel motor rated power	2 MW	Inertia constant of flywheel motor	4.7 s

.

Due to the difference of dynamic response characteristics of different energy storage technologies, their effects on power suppression under different wind speed fluctuation scenarios are slightly different. In order to study the dynamic response characteristics and effects of different energy storage technologies on suppressing wind turbine output power fluctuation, two different scenarios of wind turbine output power are obtained by changing the fluctuation degree of wind speed. Wind speed fluctuation is large in scenario 1 and small in scenario 2. The wind speeds in the two scenarios are set as follows. Scenario 1 is the output power of the wind turbine when the gust component is added based on the mean wind speed. The mean wind speed is set as 9 m/s, the duration of gust is set as 2 s to 4 s and the peak speed of gust is set as 3 m/s; Scenario 2 is the output power of wind turbine when the ramp wind component is added based on the mean wind speed. The mean wind speed is set as 9 m/s, the duration of ramp wind is set as 3 s to 4 s and the maximum speed of ramp wind is set as 3 m/s. The wind speed in two scenarios is shown in Figure 10. The above control strategy is used to suppress the fluctuation output power of wind turbine and the control objective is to stabilize the output power of PCC at 2 MW. Therefore, the reference output power of PCC is set as 2 MW, and the difference between the reference output power of PCC and the actual output power of wind turbine is used as the reference output power of the energy storage system in the simulation. The output power diagrams of PCC and energy storage system under two scenarios are shown in Figure 11 and Figure 12.

4.2. Power Response Characteristics of Three Energy Storage Technologies

Figure 11b and Figure 12b, respectively, represent the actual output power and reference output power of the energy storage system under two power fluctuation scenarios. The reference output power of the energy storage system is also called the command power of the energy storage system, which is the expected output power of the energy storage system. The value of the reference output power of the energy storage system is expressed as the difference between the reference power of PCC and the actual output power of the wind turbine. It can be seen from Figure 11b that in the period of 3 s to 3.2 s, the response of the three energy storage systems to the high-power charge and discharge instructions has different degrees of hysteresis. During this period, the power response time of capacitor energy storage is 62 ms, the power response time of battery energy storage is 107 ms and the power response time of flywheel energy storage is 133 ms. The battery energy storage system and the supercapacitor energy storage system can generally track the trend of the power command curve. In contrast, the flywheel energy storage system’s ability to track the power command curve has declined in the face of short-term high-power charge and discharge instructions due to its inertia. It can be seen from Figure 12b that in the period of 3.2 s to 3.5 s, the power response time of capacitor energy storage is 92 ms, the power response time of battery energy storage is 39 ms and the power response time of flywheel energy storage is 66 ms. Therefore, when the power fluctuation degree of wind turbine is small, the three energy storage systems can respond well to the command power curve.

4.3. Effects of Three Energy Storage Technologies on Suppressing Wind Power Fluctuation

As can be seen from Figure 11a and Figure 12a, when the amplitude of wind power fluctuation is large, the three kinds of energy storage have obvious suppressing effects on the output power of PCC. When the amplitude of wind power fluctuation is small, the three kinds of energy storage have little significant suppressing effect on the output fluctuation power.

To reflect the fluctuation of wind power accurately, a stability index is defined [22], which reveals the stability of the output power change in the given time window. The specific definition is shown as follows [25]:

$${\delta }_i=\frac{P_{i+1}-P_i}{P_{c}T}\times 100\%,\hspace{1em}i=1,2,\dots ,n-1$$

(7)

$$E_{\delta }=\frac{1}{n-1}{\displaystyle \sum _{i=1}^{n-1}\left|{\delta }_i\right|}$$

(8)

where P_i+1 is the sampling value of the data at the current time, MW; P_i is the sampling value at the previous moment, MW; P_c is the rated capacity of the wind turbine, MW; T is the sampling period of wind power change rate data, s and n is the number of data points in a given time window. The value of E_δ reflects the degree of instability of data changes. The smaller the value, the better the stability.

The sampling time is set as 0.5 s and the stability of output power of PCC is calculated before and after accessing three energy storage systems under two scenarios according to Equations (7) and (8). The results are shown in

Table 2. Comparison of suppression effect of three energy storage systems on wind power fluctuation.

Whether to Access Energy Storage System	Power Fluctuation Stability (Eδ) in Scenario 1	Power Fluctuation Stability (Eδ) in Scenario 2
Without energy storage system	0.179	0.075
Access battery energy storage system	0.065	0.0415
Access supercapacitor energy storage system	0.068	0.0410
Access flywheel energy storage system	0.055	0.0421

. It can be seen that compared with the situation without the energy storage system, the power fluctuation stability value E_δ of PCC after the three energy storage systems are accessed respectively all decrease when simulation is conducted in two scenarios. It indicates that, with the control strategy, three energy storage technologies have a suppressing effect on the output power fluctuation of the wind turbine. In Scenario 1, when the energy storage system is not accessed, the power fluctuation degree is large, and the value of E_δ is 0.179. In this case, the flywheel energy storage system has the best effect on suppressing the wind power fluctuation. After the flywheel energy storage is accessed, the value of E_δ is 0.055, which suppresses about 70% of the power fluctuation. In Scenario 2, when the energy storage system is not accessed, the power fluctuation degree is small, and the value of E_δ is 0.075. In this case, the three energy storage technologies have basically the same effect on suppressing wind power fluctuation. After the three energy storage systems are accessed, the value of E_δ is about 0.04, which suppresses about 40% of the power fluctuation.

5. Conclusions

Through the establishment of a wind storage system model, this paper simulates the dynamic response characteristics and effects of three energy storage systems on suppressing wind power fluctuation under two wind speed fluctuation scenarios. Moreover, the stability of output power is quantitatively analyzed. The conclusions are as follows:

(1) In terms of the dynamic response characteristics of the energy storage systems on suppressing wind power fluctuation, when the power fluctuation degree of wind turbine is large, the power response times of capacitor, battery and flywheel energy storage are respectively 62 ms, 107 ms, 133 ms. The flywheel energy storage system’s ability to track the power instruction curve has declined in the face of short-term high-power charge and discharge instructions due to its inertia; when the power fluctuation degree of the wind turbine is small, the three energy storage systems can respond well to the command power curve.

(2) In terms of the effects of the energy storage systems on suppressing wind power fluctuation, when wind power fluctuation is large, flywheel energy storage has a better suppression effect on wind power fluctuation, which can suppress about 70% of power fluctuation.

In this study, three kinds of energy storage systems are used to preliminarily suppress the wind power fluctuation. However, the results show that the battery and supercapacitor energy system has the state of excessive discharge. The corresponding control strategy can be improved later to make its suppression effect better.