Dynamic performance improvement of an isolated wind-diesel system with intelligently controlled supercapacitor energy storage system

Abstract

This paper aims to develop a supervisory control scheme to enhance the effectiveness and profitability of a small-rating super capacitor energy storage system (SCESS) used in load-frequency-control applications. The proposed approach, which uses one-step-ahead adaptive predictive control (APC), adeptly handles the operational limitations of the SCESS. For the purpose of online estimation of system parameters, the recursive least square (RLS) algorithm has been employed. The system can be characterized as a two input two output system, wherein the real and reactive powers required by the SCESS serve as control signals issued by the controller. The SCESS voltage is subject to constraints to limit energy trade within specific bounds. The proposed control scheme successfully maintains the voltage constraints of the SCESS while significantly reducing frequency and voltage deviations in the presence of two disturbances, viz; load disturbance and wind disturbance. The effectiveness of the proposed scheme is demonstrated through simulation experiments on an isolated hybrid wind diesel power system, which addresses several modeling and design aspects.

Keywords

Frequency control wind-diesel systems SCESS wind disturbance load disturbance MATLAB/Simulink

Introduction

Transmission of electricity to off-grid sites is one of the key problems faced by isolated communities. Because of high transmission losses, high transportation costs, sustainable supply of fuel and several other factors, the transmission of electricity to such remote locations is not only a cumbersome job but also a challenging task (Rashid et al., 2023). To meet the demand of such locations has, therefore, become the need of the hour. One of the obvious solutions to tackle this problem, provided by the researchers, is to install hybrid systems either in isolated mode or grid-connected mode. Hybrid systems are self-sufficient systems that combine renewable sources of energy with non-renewable sources of energy to maintain the reliability of the system. Due to the several advantages of wind energy, it has become the prime focus for researchers. However, the intermittent nature of wind and integration of wind energy systems through power electronic interfaces lowers the inertia of the system, thereby, adding instability in the system in terms of perturbations in power demand (Thoker and Lone, 2021).

Energy storage devices are considered as the favorable option to confront the issues of power quality. The energy storage device increases the inertia of the system, making the system immune to the perturbations like load disturbances, wind disturbances, wind park disconnection etc and hence enhancing the robustness of the system. Various energy storage technologies with distinctive features such as Superconducting magnetic energy storage (SMES), battery energy storage system (BESS), flywheel energy storage system (FESS), supercapacitor energy storage system (SCESS) etc. have been reported in the literature for tackling the intermittent nature of wind, load leveling, frequency modulation etc. Therefore, improving the dynamic behavior of wind integrated power systems can be accomplished through the implementation of an efficient method such as the addition of energy storage to the wind energy facility (Zargar et al., 2021).

Superconducting magnetic energy storage (SMES) and battery energy storage have been used to enhance the dynamic performance of power systems and for load leveling applications (Mufti et al., 2015). Unfortunately, operation of these systems and maintenance come with a number of challenges and are exorbitant. BESS has a high energy density but a low power density. However, SMES has a slightly lower storage density but is more expensive (Zargar et al., 2021). The storage density of capacitors developed with current technology is only about $10^{5} J / m^{3}$ . The dielectric losses and poor energy density of the capacitors make it less desirable as a bulk energy storage device capable of load balancing during significant incidents of generation loss. However, the power frequency and tieline power oscillations caused by small fluctuations in load can be successfully dampened by a small rating capacitive energy storage (CES) (Tripathy and Mishra, 1996; Tripathy et al., 1991). Owing to its extensive features, super capacitors have become one of the most viable energy storage technologies for load balancing and damping undesirable oscillations in the system (Mufti et al., 2015). Supercapacitors are extremely effective and durable, have no moving parts, don’t need to be heated or cooled, and are not required to experience any internal chemical changes in order to function effectively. Moreover, they basically do not need any maintenance, and they have a lifespan estimated in decades with minimal life reduction from repeated cycles. So, a suitable approach for better load frequency control (LFC) in a hybrid power system is to apply a small-rating SCESS (Mufti et al., 2015; Syed and Mufti, 2021). The introduction of renewable energy sources, particularly wind turbines, increases the difficulty of regulating frequency and tie power. As a result, a plethora of strategies for ESS control have been proposed in the literature (Nisa and Mufti, 2021; Syed and Mufti, 2021).

The frequency control of a hybrid system that has been supported with a fuzzy-controlled battery energy storage system is the focus of research by Zhang et al. (2016). Sebastián (2016), the use of a battery as an energy storage device for frequency regulation and peak shaving in a wind-diesel system is proposed. Short-term energy distributed predictive control for a wind farm is proposed by Zhao et al. (2016). Microgrid frequency regulation with flywheels and fuel cells has been investigated by Vidyanandan and Senroy (2016). The effects of load and wind power variations can be mitigated by using the discrete control scheme reported by Nisa and Mufti (2021). But the DFIG is not involved in the primary frequency support, and neither are the DC link dynamics of SCESS taken into account. Despite this, widespread implementation of these technologies is limited for a variety of reasons, including those pertaining to technology and the economy (Syed et al., 2023; Xiong et al., 2018). Optimized and parameter-identified advanced controls result in slow computation. Thus, their use in real-time applications has become cumbersome. In addition, there is no systematic approach for the design, development, and use of Fuzzy logic Control (FLC), making it highly dependent on the designer’s expertise (Ahsan and Din Mufti, 2020; Syed et al., 2023). Mufti et al. (2007) proposed fuzzy logic-based SCS for improved AGC, but it lacked the capability to maintain SCS voltage within allowable limits.

This paper presents a simulation and modeling of a hybrid wind-diesel system that integrates an adaptively controlled SCESS energy storage device. The energy storage device is located at bus 4, the load bus. Buses 1 and 2 are connected to two synchronous generators, each of which has an automatic voltage regulator and a governor.To harness wind power, an equivalent induction machine is installed at bus 3, as depicted in Figure 1. In other words, this study aims to develop a comprehensive understanding of the hybrid wind-diesel system with SCESS energy storage, which can be valuable in enhancing the system’s overall performance and efficiency. Ultimately, this study highlights the importance of integrating energy storage devices in hybrid energy systems to improve their stability and reliability.

Figure 1.

Hybrid wind-diesel-SCESS system.

Super capacitor energy storage system

A SCESS with a low rating has been incorporated in a hybrid wind diesel system in order to supplement frequency regulation in the event of any disturbance in the system. Figure 2(a) depicts the configuration of the power conversion system (PCS) that is used to interface the SCESS with the electric utility grid. This is accomplished by using a pair of power converters. The DC link voltage is supplied by the grid side converter (GSC), and a fast bidirectional interface is set up by a combined buck-boost DC-DC converter. $L_{f}$ is the converter’s filter inductance.

Figure 2.

(a) Configuration of SCESS and (b) primary control loop of SCESS.

The supervisory controller directs the primary control loop to convert power commands into current commands for the supercapacitor bank (SCB), utilizing the PI controller in the inner current control loop to regulate the flow of current through the SCB via the dc-dc converter. By switching between the two modes of operation viz. the buck or charge mode and the boost or discharge mode the bidirectional dc-dc converter facilitates energy transfer bidirectionally between the input and output sides with the chopper employing modulation of the buck/boost switch for buck/boost mode. Figure 2(b) illustrates the non-linear average value model of SCESS and its primary control (Mufti et al., 2015; Syed et al., 2023). Moreover, the DC link voltage, $V_{d c}$ , serves to express the average output voltage, $V_{S C B}$ , as:

V_{S C B} = m V_{d c}

(1)

where, $m$ , is the modulation index given by;

m = {\begin{matrix} D; buck mode \\ 1 - D; boost mode \end{matrix}

$m ϵ [0, 1]$ and $D$ is the duty ratio of the buck-boost switch dictated by the mode of operation.

Problem formulation

Controller formulation and parameter identification

An adaptive predictive control scheme has been introduced for an isolated wind-diesel power system. This scheme utilizes a one-step ahead approach based on the SCESS unit. The system has been specifically designed to proficiently manage any uncertainties that may arise within the system. It is well-suited for nonlinear and time-varying systems. In the event of any unexpected changes, the system automatically updates itself to ensure optimal performance. The advanced on-line identification technique employed for parameter estimation is depicted in Figure 3. This method generates a control signal using a pre-assigned model order that can be aptly characterized by an autoregressive discrete time model as (Mufti et al., 2015; Zargar et al., 2021):

Y (k) = \sum_{i = 1}^{n} A_{i} Y (k - i) + \sum_{i = 1}^{n} B_{i} U (k - i)

(2)

Figure 3.

Representation of TITO and MISO systems.

For a system with two inputs and two outputs, equation (2) may be expressed as follows:

\begin{matrix} Y (k) = A_{1} Y (k - 1) + A_{2} Y (k - 2) + B_{1} U (k - 1) + \\ B_{2} U (k - 2) \end{matrix}

(3)

Here, $Y (k)$ refers to an output vector that comprises the deviation variables, while $U (k)$ stands for a control vector for the $kth$ sampling instant, and $n$ signifies the model order. The model parameter matrices, including $A_{1}, A_{2}, B_{1}$ , and $B_{2}$ , necessitate online identification and are presented in matrix form below

\begin{matrix} A_{1} = [\begin{matrix} a_{11}^{1} & a_{12}^{1} \\ a_{21}^{1} & a_{22}^{1} \end{matrix}] & A_{2} = [\begin{matrix} a_{11}^{2} & a_{12}^{2} \\ a_{21}^{2} & a_{22}^{2} \end{matrix}] \\ B_{1} = [\begin{matrix} b_{11}^{1} & b_{12}^{1} \\ b_{21}^{1} & b_{22}^{1} \end{matrix}] & B_{2} = [\begin{matrix} b_{11}^{2} & b_{12}^{2} \\ b_{21}^{2} & b_{22}^{2} \end{matrix}] \end{matrix}

The main objective of the controller is to reduce the voltage and frequency deviations whenever the system is subjected to disturbances which can be achieved by defining a cost function for the controller. The purpose of the cost function is to regulate the system output and prevent the excessive control as well (Mufti et al., 1996).

The cost function is formulated as follows:

J = q_{1} {\hat{y}}_{1}^{2} (k + 1) + q_{2} {\hat{y}}_{2}^{2} (k + 1) + r_{1} u_{1}^{2} (k) + r_{2} u_{2}^{2} (k)

(4)

Here, $q_{1}, q_{2}$ represent the components of a weightage output matrix, $Q$ , whereas $r_{1}, r_{2}$ signify the constituents of a control signal matrix, $R$ . The optimal values of $r_{1}$ and $r_{2}$ are evaluated based on the system’s performance index. The matrix $[R]$ is a diagonal matrix with the values of $r_{1} = 0.7$ and $r_{2} = 2.7$ .

The performance index is calculated as:

P_{ISE} = W_{1} \int_{0}^{14} {(Δ f)}^{2} dt + W_{2} \sum_{i = 1}^{2} \int_{0}^{14} {(Δ | V_{i} |)}^{2} + W_{3} \int_{0}^{14} {(Δ | V_{3} |)}^{2} dt + W_{4} \int_{0}^{14} {(Δ | V_{4} |)}^{2} dt

(5)

Where, $W_{1}$ , $W_{2}$ , $W_{3}$ , and $W_{4}$ are the weights corresponding to the various variable.

The control law is obtained by setting equation (4) to zero (Zargar et al., 2021). For formulation of optimal control law, expression for $Y (k + 1)$ is required which is obtained by writing (3) for the $(k + 1)$ th sampling instant in terms of $y_{1}$ and $y_{2}$ as:

\begin{matrix} {\hat{y}}_{1} (k + 1) = a_{11}^{1} y_{1} (k) + a_{12}^{1} y_{2} (k) + a_{11}^{2} y_{1} (k - 1) + \\ a_{12}^{2} y_{2} (k - 1) + b_{11}^{1} u_{1} (k) + b_{12}^{1} u_{2} (k) + \\ b_{11}^{2} u_{1} (k - 1) + b_{12}^{2} u_{2} (k - 1) \\ {\hat{y}}_{2} (k + 1) = a_{21}^{1} y_{1} (k) + a_{22}^{1} y_{2} (k) + a_{21}^{2} y_{1} (k - 1) + \\ a_{22}^{2} y_{2} (k - 1) + b_{21}^{1} u_{1} (k) + b_{22}^{1} u_{2} (k) + \\ b_{21}^{2} u_{1} (k - 1) + b_{22}^{2} u_{2} (k - 1) \end{matrix}

(6)

Minimizing (4) with respect to $U (k)$ will give the optimal control law, so setting both $\frac{\partial J}{\partial u_{1} (k)}$ and $\frac{\partial J}{\partial u_{2} (k)}$ to zero gives.

2 q_{1} {\hat{y}}_{1} (k + 1) \frac{\partial {\hat{y}}_{1} (k + 1)}{\partial u_{1} (k)} + 2 q_{2} {\hat{y}}_{2} (k + 1) \frac{\partial {\hat{y}}_{2} (k + 1)}{\partial u_{1} (k)} + 2 r_{1} u_{1} (k) = 0

(7)

\begin{matrix} 2 q_{1} {\hat{y}}_{1} (k + 1) \frac{\partial {\hat{y}}_{1} (k + 1)}{\partial u_{2} (k)} + 2 q_{2} {\hat{y}}_{2} (k + 1) \frac{\partial {\hat{y}}_{2} (k + 1)}{\partial u_{2} (k)} + \\ 2 r_{2} u_{2} (k) = 0 \end{matrix}

(8)

Expanding (7) and (8) by substituting $y_{1} (k + 1)$ and $y_{2} (k + 1)$ and rearranging the terms leads to a compact form of matrix equation expressed as:

\begin{matrix} U (k) = - {[B_{1}^{T} Q B_{1} + R]}^{- 1} B_{1}^{T} Q [A_{1} Y (k) + A_{2} Y (k - 1) + \\ B_{2} U (k - 1)] \end{matrix}

(9)

Equation (9) in a general form can be written as (10) which is the control law for achieving an optimal control.

U (k) = - {[B_{1}^{T} Q B_{1} + R]}^{- 1} B_{1}^{T} Q [\sum_{i = 1}^{n} A_{i} Y (k - i) + \sum_{i = 2}^{n} B_{i} U (k - i)]

(10)

Two MISO approach is preferred for n-step ahead identification, with TITO system splitted into two MISO system as shown in Figure 4.

Figure 4.

Representation of (a) TITO and (b) MISO systems.

The application of optimal control signal $u (k)$ is limited to instances where the model parameters $(a_{i}, b_{i})$ are known. In order to estimate these parameters at every sampling instant, a recursive least square algorithm is employed. To facilitate this estimation, the discrete time model of (2) is reformulated as:

x (k) = h^{T} (k) θ (k - 1)

(11)

where $x (k) = y (k)$

h^{T} (k) = [y (k - 1), \dots y (k - n), u (k - 1), \dots . u (k - n)]

(12)

θ^{T} (k - 1) = [a_{1}, \dots \dots a_{n}, b_{1}, \dots, b_{n}]

(13)

The recursive algorithm for parameter estimation is expressed through a set of equations as (Mufti et al., 1996):

\hat{θ} (k + 1) = θ (k + 1) + K (k + 1) e (k + 1) θ

(14)

K (k + 1) = P (k) h (k + 1) / [β + h^{T} (k + 1) P (k) h (k + 1)]

(15)

P (k + 1) = [I - K (k + 1) h^{T} (k + 1)] P (k) / β

(16)

e (k + 1) = x (k + 1) - h^{T} (k + 1) \hat{θ} (k)

(17)

where the caret denotes the estimated values and $β$ satisfies the condition of being greater than zero. The aforementioned approach provides an effective means of continuously estimating the model parameters, thus allowing for the application of optimal control signal $u (k)$ in real-time scenarios. Overall, the use of a recursive least squares algorithm for parameter estimation is a crucial aspect of achieving optimal control in complex systems.

Constraints on SCESS energy level

The power command for a SCESS device must always be within the limits of the converter rating. In order to adhere to the prescribed energy exchange limits between the wind-diesel system and the SCESS unit, constraints are placed on the energy transfer by the power conversion system. A minimum stored energy level must be enforced on the SCESS unit for continuous control (Iqbal et al., 2009; Zargar et al., 2021). To ensure profitable operation and uninterrupted control, the voltage/energy of the SCESS must always remain within the upper and lower limits. By discretizing time into discrete intervals, the discrete-time model is used to represent the relationship between the energy level and the voltage of the storage unit as depicted in Figure 5, which takes into account the time constant of the SCESS unit $(T)$ , the zero-order hold $(ZOH)$ , and the steady-state and actual energy levels ( $E_{SCESS}^{0}$ and $E_{SCESS}$ , respectively) to make energy level predictions. Online constraints on energy/voltage must be imposed to ensure optimal operation.

Figure 5.

Prediction for energy level of SCESS.

An imperative factor that holds great significance in the realm of supercapacitors is the discharge factor $d$ . It indicates the range of usable voltage for discharging the stored energy and is the ratio of the minimum voltage $v_{m}$ , below which the SCB cannot be discharged, to the maximum acceptable voltage $V_{M}$ . Evidently, the entire energy $W_{M}$ , within the supercapacitor is not utilized in order to optimize the efficiencies. Rather, only a proportion, referred as the useful energy $W_{U}$ , can be harnessed (Iqbal et al., 2009). Subsequently, it is noteworthy that the useful energy that can be extracted from the supercapacitor is limited to 75% of the total energy, which restricts the discharge factor to 50%. As a result, the voltage of the supercapacitor, or stored energy, must be prohibited from exceeding specific upper and lower boundaries, so as to prevent any unwanted outcomes. The energy stored by SCESS at the $(k + 1)$ sampling instant can be correlated to the power command at the $kth$ sampling instant through the utilization of the discrete time system (Zargar et al., 2021). The pulse transfer function, as depicted in Figure 5, can be expressed as follows.

G (z) = \frac{a z^{- 1} + b z^{- 2}}{1 + c z^{- 1} + d z^{- 2}}

(18)

where $a, b, c$ and $d$ represent the constants for time constant T.

For $(k + 1) th$ sampling instant, equation (2) is given as:

E_{SCESS} (k + 1) + c E_{SCESS} (k) + d E_{SCESS} (k - 1) = a P_{SCESS} (k) + b P_{SCESS} (k - 1)

(19)

P_{SCESS} (k) = \frac{E_{SCESS} (k + 1) - b P_{SCESS} (k - 1) + c E_{SCESS} (k) + d E_{SCESS} (k - 1)}{a}

(20)

In order to ensure that the energy limits are not surpassed, as outlined by equations (5) and (6), a judicious implementation of the power directive at the $kth$ sampling instance is imperative (Mufti et al., 1996; Zargar et al., 2021)

P_{SCESS} (k)_{\min} = \frac{E_{SCESSmin} + [- b P_{SCESS} (k - 1) + c E_{SCESS} (k) + d E_{SCESS} (k - 1)]}{a}

(21)

P_{SCESS} (k)_{\max} = \frac{E_{SCESSmax} + [- b P_{SCESS} (k - 1) + c E_{SCESS} (k) + d E_{SCESS} (k - 1)]}{a}

(22)

To this end, a Simulink model has been developed, that imposes these constraints on the SCESS energy, based on the equations (2)–(6).

Implementation of proposed scheme on SCESS unit

The present section encompasses the execution of the control and identification algorithm derived in section “Controller formulation and parameter identification.” The SCESS unit under consideration is a two input two output system, where $y_{1}$ and $y_{2}$ represent the regulated variables, as defined in (7) and (8), and $u_{1}$ and $u_{2}$ denote the active power $(P_{d})$ and reactive power $(Q_{d})$ command signals dispensed by the regulator to the SCESS unit.

y_{1} = Δ f + k_{c} (V_{SC}^{0} - V_{SC})

(23)

y_{2} = Δ V_{4}

(24)

Specifically, $y_{1}$ is a function of SCESS voltage and frequency deviations, while $y_{2}$ represents the load bus voltage deviation of the wind-diesel system. Assuming the rated system frequency to be $f_{o}$ , system frequency deviation is articulated as;

Δ f = [(\frac{1}{2 π} ω_{COI}) - f_{o}]

(25)

For continuous control operation, the second term in (19) is incorporated, enabling the SCESS voltage to restore to its nominal value after tackling a disturbance. In the condition of a steady state, it is observed that both the frequency and SCESS voltage deviations are rendered null. It is imperative to exercise judiciousness in selecting the value of $k_{c}$ , as a large value of $k_{c}$ may impede the efficacy of SCESS in mitigating the frequency deviations, while a very small value of $k_{c}$ may not suffice in restoring the SCESS voltage to its initial value (Mufti et al., 1996; Zargar et al., 2021).

Time domain simulation results

The hybrid wind diesel test-bed system is initially presumed to be in a steady state. However, upon being subjected to a combination of load and wind disturbances, this balance is disrupted, leading to a deviation from its steady state values. To augment the system power quality under such scenarios, an adaptively controlled SCESS unit is proposed.

Load disturbance

Due to the low inertia of the systems, load perturbation is a typical sort of perturbation that appears in hybrid wind power systems. As the bus admittance matrix adapts to the new load requirement, the system experiences frequency and voltage fluctuations (Rashid et al., 2023; Thoker and Lone, 2021). We assume that the system’s initial load was 260 $kW$ , which equates to 182 $kVAR$ . The active/reactive power load disturbance alters the demand for load and the load profile, as shown in Table 1.

Table 1.

Load description for an alteration in reactive/active load power.

Time (seconds)	Active power (kW)	Reactive power (KVAR)
$0 < t < 10$	260	182
$10 < t < 20$	234	163.8
$20 < t < 30$	260	182

Figures 6 and 7 show the simulation findings for this case study. The results show that SCESS is able to keep the system power quality dynamics stable by lowering the maximum voltage and frequency fluctuations.

Figure 6.

System frequency deviation trajectory under load disturbance.

Figure 7.

Dynamic power system and SCESS responses for load disturbance.

Wind disturbance

The second scenario for which the developed strategy has been assessed is wind disturbance. To determine how effectively SCESS improves the system power quality, the test-bed system is operated for a total of 14 seconds under wind turbulence. In the beginning, the wind turbines are only producing 36 kW of power, but it is anticipated that the load demand will be 260 kW and 182 kVAR. For $t > t_{0}$ , wind turbines must operate in a random wind profile. Figure 9 depicts the wind power outline for this instance. The system dynamic responses for this case study are shown in Figures 8 to 10 below.

Figure 8.

System Frequency deviation profile for wind disturbance.

Figure 9.

(a) Wind speed trajectory and (b) wind power output.

Figure 10.

Dynamic power system and SCESS responses for wind disturbance.

The simulation results are summarized as follows:

The application of the proposed framework not only leads to the elimination of persistent undesirable oscillations but also results in a notable reduction in both frequency and voltage deviations as illustrated in Figures 6 –8 and 10 for both cases respectively. Through the utilization of the proposed control scheme to regulate the SCESS, a significant decrease in the frequency and voltage deviations is observed. This demonstrates that the suggested framework is quite effective at containing peak frequency and voltage variations.

The proposed methodology compels the SCESS to function in close proximity to its limitations, thereby successfully and efficiently mitigating power-system oscillations.

The SCESS converter is consistently able to manage a power that never surpasses the rated value of the converter, equivalent to 0.07 p.u. of power.

The voltage of the SCESS consistently sustains itself within the permissible limits, both in the upper and lower thresholds.

After addressing a perturbation, the SCESS obtains its designated voltage level, thereby accumulating energy and persisting in a prepared state to confront additional load disruptions.

From the obtained results, it can be inferred that the adaptive predictive controller is proficient in generating the dynamic power constraints with satisfactory outcomes. This establishes the justification for the utilization of the aforementioned controller in the SCESS based wind diesel power system.

Conclusion

This paper presents the application of a one-step ahead adaptive control scheme based on a SCESS unit to mitigate power quality issues caused by the random nature of wind. A discrete-time model of the storage unit is utilized to generate a suitable power command for the unit. System parameters are estimated using a two-input, two-output system that is split into two two-input, single-output systems. The recursive least square algorithm is used for parameter estimation and a separate model is applied to impose constraints on the SCESS energy/voltage level. By incorporating these constraints into the controller formulation, significant reductions in frequency and voltage deviations are observed, and the SCESS voltage/energy limits are always within the prescribed limits. The proposed methodology has undergone testing for disturbances belonging to two distinct categories, namely continuous wind perturbation and higher magnitude load disturbance. Despite the presence of such disturbances, the SCESS constraints are always upheld while simultaneously minimizing the deviations in frequency and voltage. The adaptive controller-based SCESS unit, as conferred in this study, has the potential to be a more favorable choice for power systems that rely on renewable sources. The selection of the regulated parameter under consideration compels the SCESS to operate as a highly efficient damping mechanism, thereby augmenting the efficacy of Load Frequency Control (LFC).

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Gazala Rashid

References

Ahsan

Din Mufti

(2020) Sweeping power system stabilisation with a parametric fuzzy predictive control of a generalised energy storage device. IET Generation Transmission & Distribution 14(25): 6087–6096.

Iqbal

Mufti

Lone

, et al. (2009) Intelligently controlled superconducting magnetic energy storage for improved load frequency control. International Journal of Power and Energy Systems 29(4): 241.

Mufti

Iqbal

Lone

, et al. (2015) Supervisory adaptive predictive control scheme for supercapacitor energy storage system. IEEE Systems Journal 9(3): 1020–1030.

Mufti

MUD

Balasubramanian

Tripathy

(1996) Self tuning control of wind-diesel power systems. In: Proceedings of international conference on power electronics, drives and energy systems for industrial growth, pp.258–264, vol. 1. New York, NY: IEEE.

Mufti

MUD

Lone

Iqbal

, et al. (2007) Improved load frequency control with superconducting magnetic energy storage in interconnected power systems. IEEJ Transactions on Electrical and Electronic Engineering 2(3): 387–397.

Nisa

Mufti

MU-D

(2021) Constrained discrete mode control of supercapacitor energy storage system for improved agc of a multi-area power system with effects of wind power. International Journal of Power Electronics 14(3): 386–403.

Rashid

Lone

Mufti

MU-D

(2023) Modeling and performance assessment of an isolated wind-diesel system with flywheel energy storage system. Wind Engineering 47: 597–606.

Sebastián

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

Syed

Mufti

MUD

(2021) Constrained neural adaptive predictive control of smes for dynamic performance improvement of power systems. Wind Engineering 45: 1531–1543.

10.

Syed

A U M

Mufti

(2023) Coordinated control of wind farm and supercapacitor energy storage system for dynamic performance reinforcement of multi-area power systems. International Journal of Power Electronics 17(3): 261–279.

11.

Thoker

Lone

(2021) Voltage and frequency control of wind–diesel power system through adaptive sliding mode control of superconducting magnetic energy storage. Wind Engineering 45: 1057–1071.

12.

Tripathy

Balasubramanian

Chandramohanan Nair

(1991) Small rating capacitive energy storage for dynamic performance improvement of automatic generation control. IEE Proceedings C 138(1): 103–111.

13.

Tripathy

Mishra

(1996) Dynamic performance of wind-diesel power system with capacitive energy storage. Energy Conversion and Management 37(12): 1787–1798.

14.

Vidyanandan

Senroy

(2016) Frequency regulation in a wind–diesel powered microgrid using flywheels and fuel cells. IET Generation Transmission & Distribution 10(3): 780–788.

15.

Xiong

Zhu

, et al. (2018) Coordinated control schemes of super-capacitor and kinetic energy of dfig for system frequency support. Energies 11(1): 103.

16.

Zargar

Mufti

MUD

Lone

(2021) Dynamic performance improvement of isolated power system using intelligently controlled smes. IET Generation Transmission & Distribution 15(3): 408–419.

17.

Zhang

Mishra

Shahidehpour

(2016) Fuzzy-logic based frequency controller for wind farms augmented with energy storage systems. IEEE Transactions on Power Systems 31(2): 1595–1603.

18.

Zhao

Guo

, et al. (2016) Optimal active power control of a wind farm equipped with energy storage system based on distributed model predictive control. IET Generation Transmission & Distribution 10(3): 669–677.