Journal of Modern Power Systems and Clean Energy

ISSN 2196-5625 CN 32-1884/TK

网刊加载中。。。

使用Chrome浏览器效果最佳,继续浏览,你可能不会看到最佳的展示效果,

确定继续浏览么?

复制成功,请在其他浏览器进行阅读

A Review on Robust and Adaptive Control Schemes for Microgrid  PDF

  • Swagat Kumar Panda (Student Member, IEEE)
  • Bidyadhar Subudhi (Senior Member, IEEE)
the School of Electrical Sciences, Indian Institute of Technology Goa, Goa, India,

Updated:2023-07-25

DOI:10.35833/MPCE.2021.000817

  • Full Text
  • Figs & Tabs
  • References
  • Authors
  • About
CITE
OUTLINE

Abstract

Several challenges are encountered while integrating microgrids (MGs) with an existing utility grid. These are low inertia, intermittent nature of renewable energy resources (RESs), sensor and actuator faults, unbalanced and nonlinear loads, supply-demand mismatch, and uncertain switching functions of the power electronic converter. MG control is fully dependent on the communication network, which is also susceptible to different types of failures such as noise in communication links, communication delay, limited bandwidth, packet dropout, and various malicious cyber attacks. Very few papers are available in the literature that focus on the review of control schemes for MG. Consequently, this paper presents a comprehensive review of different robust and adaptive control schemes that address the challenges encountered due to communication constraints, uncertainties, and disturbances for different MG topologies such as AC, DC, and hybrid AC/DC MGs, as well as the current research trends in the field of robust and adaptive control. It can be concluded that to achieve control objectives of MG and overcome the existing challenges, robust and adaptive controllers show significantly improved performance in terms of transient and steady-state behavior and robustness as compared to traditional controllers.

I. Introduction

THE microgrid (MG) is a small-scale power system network suitable for power supply to remote areas and provides flexibility to the existing network. An MG consists of several distributed energy resources (DERs), power electronic converters, energy storage elements, loads, and control units based on communication topologies. It has several advantages over a traditional grid, such as reduction in power losses, voltage drop during transmission, reduction in carbon dioxide emission, low installation cost, reliable and uninterrupted power supply, a two-way controlled power flow, and improvement of power quality [

1].

The traditional grid consists of a number of synchronous generators which provide inertia to the power system network to control voltage and frequency. However, in MG, the renewable energy resource (RES) based distributed generator (DG) replaces the synchronous generator, which does not provide inertia to the system. So, the implementation of conventional droop control in power system is easier as compared to that in MG. To control voltage and frequency in MG, active power-frequency (P-ω) and reactive power-voltage (Q-V) based droop control mimics the behavior of synchronous generator in the traditional grid. In the traditional grid, there exists centralized generation, whereas in case of MG, DG exists. Therefore, the centralized supervisory control scheme cannot handle the overall burden of DG because of its complex network and single-point failure. The traditional grid adopts manual restoration during fault in the system with unidirectional communication, whereas MG has self-healing and self-monitoring facilities provided by advanced control with bidirectional communication network. The conventional proportional-integral (PI) controller adopted by the traditional grid cannot provide enough robustness during MG operations, i.e., plug and play (PnP), and the presence of uncertainties arises from system modelling or external environment.

However, integration of MG with a utility grid poses several challenges such as communication constraints, input and load uncertainties, unmodeled dynamics, disturbances, and bidirectional power flow [

2]. These affect the dynamics of MG and its transient response, resulting in degraded power quality and stability. Efficient control of an MG necessitates an effective communication network. Various faults, unwanted noise injected by malicious cyber attacks, disturbances, failure of communication links, and increasing communication delay hinder towards achieving the reliable operation and control of MG [3].

Other factors such as PnP operation, data volume, size of the MG network also affect the effective communication. To resolve these issues, a suitable controller needs to be designed, which not only provides stability but also facilitates restoration of the normal operation of the MG system by regulating voltage and frequency at nominal values, sharing accurate active and reactive power, maintaining optimum power quality, and grid synchronization.

Several control schemes have been reported in the recent literature [

4], [5] for AC, DC, and hybrid AC/DC MGs. Conventional linear controllers may fail in providing desired performance requirements, especially when changes in the system and environment occur during the operation. In recent years, distributed adaptive and robust controls have gained immense popularity. Adaptive and robust control techniques or the combination of both are adopted for obtaining improved performance to restore MG objectives despite the aforesaid challenges in MG. Table I shows the comparison between conventional controllers and robust and adaptive controllers.

Table I  Comparison Between Conventional Controllers and Robust and Adaptive Controllers
CharacteristicConventional controllerRobust and adaptive controller
Complexity Simple Complex
Robustness Low High
Scalability Moderate High
Adaptability Low High
Gain Fixed Dynamic
Reliability Low High
Optimal operation Not guaranteed Guaranteed
Dependency over system dynamics More Less

Combining robust and adaptive control schemes with traditional controllers adds to the merits of traditional controllers to achieve robustness. It includes robust hierarchical control, robust sliding mode control, robust H control, robust backstepping control, robust adaptive control, adaptive droop control, adaptive sliding mode control, adaptive backstepping control, adaptive event-triggered control, and adaptive passivity-based control. There is a significant increase in dynamic and transient responses of the system by employing one of these control schemes. Since the last two decades, there has been a significant progress in the development of MG controllers to deal with different challenges. Therefore, this paper presents the various challenges and a comprehensive review of the robust and adaptive control schemes for different MG topologies (AC, DC, hybrid AC/DC) and different operation modes (grid-connected and islanded).

The rest of this paper is organized as follows. In Section II, a brief discussion on challenges in MG operation and control is provided. Section III discusses control issues associated with MG in term of robustness. An overview of robust and adaptive control techniques is presented in Section IV. Section V presents the future trends in the robust and adaptive control. Section VI provides the conclusion.

II. Challenges in MG Operation and Control

MG is equipped with numerous power electronic converters, as well as communication and control protection devices to achieve optimum control performances as compared to traditional grid. Integration of MG into the existing power system brings several functional edges: reduction in carbon emission, continuous and independent supply of power to local and remote areas, improved overall power system quality, provision of backup power source during blackout, provision of PnP operations, and efficient bidirectional power flow. Although this integration brings several benefits, some challenges such as control, power quality, supply dependability, and outage resynchronization time are encountered.

Figure 1 shows a generic model of MG, where VSI, SOC, and PCC stands for voltage source inverter, state of charge, and point of common coupling, respectively. The challenges encountered in the MG operation and control are briefly discussed below.

Fig. 1  A generic model of MG.

A. Intermittent RESs

One of the advantages of MG is that it facilities RES integration on the generation. However, several factors affect renewable energy generation.

In case of photovoltaic (PV) array, the availability of sunlight, duration of the day time, partial shedding, solar insolation over PV array, and rain hinders the optimal generation of electricity [

2]. Similarly, in the case of wind generators, the wind velocity varies intermittently with respect to the season and altitude. As a result, the voltage fluctuations, flickers, and issues related to system stability are the common problems for such resources, which affect the dynamic and transient performance of MGs.

In an MG, several DGs are added to meet the increasing load demand. An MG is encountered with various uncertainties and nonlinearities such as pulse width modulation (PWM) switching function of VSI [

6], intermittent in DGs, and demand consumption.

B. Fault

A fault occurs in MGs in many ways. It can be either through any hardware damage such as sensor and actuator faults or by means of software owing to the fault in communication links and malicious attacks. These faults adversely affect the performance of MG operation, and lead to unstable operation of MGs.

1) Sensor and Actuator Faults

MG consists of various electrical and electronic components that are more susceptible to various types of faults. It includes changes in physical parameters due to adverse environmental conditions, electromagnetic compatibility problems, loose connection, and contact problems. Figure 2 shows the topology of an AC MG. The sensors collect the VSI outputs, i.e., three-phase AC voltage and current vi and ii and LC filter outputs voi and ioi. vodi, voqi and iodi, ioqi are the direct- and quadrature-axis components of voi and ioi of the ith DG, respectively. The communication channel is established between the ith and jth DGs. On the load side, AC load appliances, the utility grid, and electric vehicles are connected through PCC. Current transformer and potential transformer collect the currents and voltages. However, if there is an error in measurement or any malfunctioning associated with the sensor device, it results in uncertainty in the system dynamics.

Fig. 2  Topology of AC MG.

The fault in power modules can be broadly classified as catastrophic fault (triggered by a single over-stress event such as over-current, over-voltage, or over-temperature rise) and wear-out fault (gradual degradation over time). The error that arises due to fault in sensing devices can be represented as additive faults. The system output can be represented as [

7]:

yif(t)=Cixi(t)+fis(t) (1)

where yif(t) is the MG output under sensor fault; Ci is the output matrix with the same dimension as the MG output; xi(t) is the system parameter matrix; and fis(t) is the sensor fault result of the ith DG. This sensor fault may be the combination of the abrupt fault caused by periodic fault and the incipient fault caused by precision degradation. Sometimes the sensors show null values due to the failure of supply. In [

8], the voltage, frequency, and active power can be represented as the combination of sensor and actuator faults. In [9], the dynamics of the ith DG with sensor faults and nonlinearities is represented as:

v˙odi,1=vodi,2v˙odi,2=(1-ϕi(t))vni+ψi(t)+di(vodi,1,vodi,2,t) (2)

where vodi,1 and vodi,2 are the filter output voltages along the direct axis; di(vodi,1,vodi,2,t) is the unmodeled dynamics of the DG, which includes unknown disturbances and uncertainties; vni is the virtual control input to DG; and ϕi(t) and ψi(t) are the biased fault severity and the partial loss of effectiveness fault severity of the ith DG, respectively.

2) Communication Link Faults

The performance of the distributed controller is based on data exchange among the neighbors. While in the cyber layer, there are possibilities of cyber attacks, which include feeding wrong data to the system, viruses, or breaking the communication links among the DGs. This makes the system vulnerable to operating abnormally. The effect of the unbounded attack on MGs is shown in [

10]. The different types of cyber attacks are explained in [11].

C. Unbalanced Conditions of MGs

Generally, the loads in MGs are unpredictable. With an increase in the penetration of inverter-interfaced DG, presence of nonlinear loads, uneven distribution of single-phase load, asymmetric line impedance, and increasing load demand in MGs, voltage unbalance generally occurs, which affects the power quality issues [

12]. This results in de-rating, mechanical stress, overheating, and decrement in a lifetime, production of ripples at a double frequency of power oscillation in the DC-link voltage, and an increase in the production of reactive power. Nonlinear and uncertain loads inject negative-sequence voltage drop across the line impedance. As a result, the MG voltage becomes unbalanced, making the system poorly damped and have loss-less energy dissipation across the converter [13]. Also, the supply/demand imbalance, utility grid failure [14], resonance phenomenon [15], and clock drift [16] affect the normal operations of the MG.

D. Disturbances

In [

3], the communication constraints are roughly categorized as time-varying network topology, fading network channels with noise disturbances, limited communication bandwidth, transmission/communication delay, uncertain links, and cyber security.

1) Communication Degradation Process

Noise, packet loss, delay, communication failure, communication uncertainties, switching or unreliable communication networks, time-varying delays, limited communication bandwidth, and additive noise in a network are the primary reasons for the communication degradation. The effect of noise on the DG output can be expressed as [

12]:

v˜odi(t)=vodi(t)+δi(t) (3)

where v˜odi(t) is the voltage perturbed with noises of the ith DG; vodi(t) is the actual voltage output of the ith DG; and δi(t) is the zero-mean Gaussian noise of the ith DG.

In [

13], the system noise is expressed in the form of a constant additive term. A state dependent stochastic noise is encountered in [14]. It is mainly due to the non-fixed measurement precision of sensors, existing uncertainties, and time-varying stochastic disturbances. Based on the adverse effects of communication noise, the system noise can be classified as additive noise (external disturbance modeled by additive Gaussian white noise) and multiplicative noise (internal dynamic disturbance coupled with the system state) [3].

2) Communication Delay

Communication delay is one of the most critical issues in the MG control. However, in most research, the delay is not considered. Some delay exists between transmitted and received signals, which is affected by the sampling rate. The delay can be classified as continuous time delay and random time delay [

14]. The continuous time delay is further categorized as propagation delay, transmission delay, processing delay, and queuing delay.

Figure 3 presents the signal flow in a communication network with different types of communication delay, where vodi(t+T) and vodi(t+2T) are the output voltages of the ith DG at T and 2T instants, respectively; td is the delayed time to reach the information at a particular instant; ts is the time instant at which MG controllers update their inputs; and θt is the communication delay. The varying time delay is td(t)=t-ts+T at receiver 1, whereas it is a constant time delay at receiver 2.

Fig. 3  Signal flow in communication network with different types of communication delay.

Besides these factors, the computation time for the generation of control inputs, limited communication speed, extra time for the measurement message reception, and execution time for the inputs contribute to the communication delay. In [

16], [17], the influence of delay on system parameters is presented. In addition to the aforementioned factors, packet dropouts prevent the information from being transferred to the intended location despite the fact that it is there in the network bandwidth. Thus, it affects the MG operation adversely. These issues need to be considered when designing controllers for an MG.

III. Control Issues Associated with MG in Term of Robustness

MG is a complex system with a sustainably large number of components. Hence, several control issues involved in the MG are presented in the following.

A. Synchronization of System Parameters Under Communication Delay

Communication constraints such as communication delay and packet loss are often encountered in the communication network [

13], which results in the time synchronization loss of system parameters and ultimately degrades dynamic performance or even causes instability to the system. The addition of time delay to the system results in shifting of system poles towards the right half of the s-plane, indicating an unstable system. Thus, for the system with time delay, it is necessary to determine the delay margin. Any delay exceeding the delay margin will make the system unstable. In [16], the delay between the local feedback measurements may shift the output voltage phase and magnitude of VSI and significantly reduce its output power. Hence, it is challenging to consider communication delay as another system parameter during the control law implementation [3].

B. Robustness Against Disturbances and Switching Topologies

The DGs are connected to the common bus via various power electronic converters. The system inertia decreases significantly. Therefore, the importance of system stability and dynamics with regard to disturbances is a major concern. The presence of disturbances in the form of component mismatch, numerical errors, and transient power oscillation results in obtaining erroneous state information. Arbitrary switching [

18] could also lead to large state transients at switching time instants. Cyber attack also affects the system through the communication channel during switching operations.

C. Seamless Switching Operation Between Grid-connected and Islanded Modes

The switching operation adds flexibility to the MG. The effective utilization of switching operations can significantly improve the system performance of MG. It can be accomplished by reducing network losses, isolating fault area, optimizing bus voltage profile, and enhancing resilience [

19], but it becomes challenging when the system is loaded to its maximum. The relative stability of the system is impacted by transient voltages and currents during the transition between grid-connected and islanded modes of MG [20].

D. Implementation of Cyber Security

The cyber system should guarantee that the data are timely accessible. Attackers block or delay the data communication. The low inertia of MG affects the transient- and steady-state stability of MG. Furthermore, in the hybrid AC/DC MG, any cyber attack in either AC or DC MG will affect the other side through AC/DC MG interlinking power converters. It deteriorates the MG functionalities, i.e., current sharing and state estimation [

3]. Therefore, to effectively defend against those cyber attacks and protect the security and integrity of the system is a critical issue. It is challenging to merge machine learning and artificial intelligence to build a data-driven cyber-enabled MG model.

E. Determinating Accurate State Information of Uncertain Model

The state information of MG is required in order to establish the operating condition of the system. As a result, MG can be properly monitored and managed. Cyber assaults tamper with sensor data, leading to errors in the calculation of state variables such as voltage and frequency [

21]. This has a negative impact on the operation of MG. As a result, despite the existence of a malicious cyber assault, it is critical to ascertain accurate state information in order to achieve MG objectives [22].

F. Maintaining Frequency and Voltage Stability and Power Quality with Respect to Challenges in MGs

The stability issues that commonly occur in MGs involve frequency and voltage stability. It indicates the ability of MG system to maintain its operation parameters within an acceptable range. The voltage and frequency stability impacts a short-term and/or long-term stability problem. The high penetration of RES reduces system inertia, leads to significant rise in the rate of change of frequency, and increases the value of nadir frequency and the potential of unstable frequency in a power system. A sudden change in loads, presence of constant power load (CPL), penetration of DG units, and supply/demand imbalance are the major causes of voltage and frequency instability.

The grid and the connected loads are designed to operate at specified frequencies with a rated voltage and current level [

21]. Any power quality problem such as voltage sag and swell, flicker, current imbalance, and current and voltage harmonics can affect the synchronization of PV system with the utility grid. Hence, voltage imbalance, harmonic content, increased reactive power demand, and frequency deviation are the foremost power quality hitches that affect the utility grid [22].

G. Effective Handling of Nonlinear Load

The presence of nonlinear load and switching resonance in the MG distorts the voltage and current harmonics. Switching operation also often introduces harmonics in the system parameters. The voltage and current include high-frequency harmonic components [

23]. Hence, in an MG, the mitigation of PCC voltage harmonics, DG line current harmonics, local load harmonics, voltage and current harmonics in the critical bus of multi-bus MG, current harmonics from individual equipment in the MG and the minimization of voltage and current harmonics of all buses in the multi-bus MG are crucial issues [24].

IV. Overview of Robust and Adaptive Control Techniques

Uncertainties arise owing to the inaccuracy when modeling a system. A suitable robust controller needs to be designed to achieve robustness. Traditional controllers cannot handle the nonlinearities and uncertainties of the MG system to achieve MG control objectives, i.e., voltage and frequency restoration and proportional active and reactive power sharing.

Robust control refers to the control of plants with inaccuracy in dynamics subject to unknown disturbances. Figure 4 shows the structure of the robust controller.

Fig. 4  Structure of robust controller.

The uncertainties arise from unmodelled dynamics of the plant, external disturbances, and unknown noisy sensor data input. The presence of such uncertainties in the MG involves uncertain PWM switching function, unknown load dynamics, intermittent RES input, and noisy sensor and actuator data that can affect the observability, controllability, and stability of the system using limited information. It does not necessitate prior knowledge of uncertainties and disturbances. Although it is advantageous in disturbance rejection and setpoint tracking, it is not a suitable option for a system that encounters rapid change of uncertainties. The review of robust control schemes is discussed in [

25].

Adaptive control is beneficial for the system that encounters parametric uncertainties and unpredictable parameter variations of system dynamics. Hence, adaptive control modifies its behavior in response to the changes in the dynamics of the system through a parameter adjustment mechanism, which is represented through dotted line in the Fig. 5. Implementation of adaptive control is easy and gives fast responses. However, it requires accurate formulation of the reference model, whose response is matched with the considered system.

Fig. 5  Structure of typical adaptive controller.

To accomplish global voltage restoration, robust sliding-mode controllers are proposed against multiple communication delays and modeled uncertainties in [

26]-[28]. These controllers rely on the upper bound of the uncertainties rather than exact information regarding uncertainty or delays.

Similarly, to achieve robustness against parametric uncertainties, exogenous disturbances, and fixed time delay in the MG, adaptive control is used to achieve voltage, frequency, and proportional active and reactive power sharing of the MG in [

29]-[31].

Both robust control and adaptive control provide robustness in the face of uncertainties and disturbances. They are also integrated with other control schemes to achieve better performance of the system. Figure 6 shows the evolution of robust and adaptive control strategies, where LQR, LMI, LQG, SISO, and MIMO stand for linear quadratic regulator, linear matrix inequality, linear quadratic Gaussian, single-input single-output, and multi-input multi-output, respectively. The detailed evolution of robust controller and adaptive controller can be found in [

32], [33].

Fig. 6  Evolution of robust and adaptive control strategies.

A. Common Robust and Adaptive Control Scheme in MG

1) Robust Hierarchical Control

The control objectives of MG vary with regard to significances and time scales. The hierarchical control handles MG operations at different control hierarchies. The communication network mainly incorporates in these hierarchies for optimized power sharing with system parameter restoration and resynchronization. Depending upon the sharing of information over communication links, the secondary control structure is classified as centralized, decentralized, and distributed. The centralized controller has the disadvantages such as single-point failure, computationally expensive, time-consuming, extensive use of communication network, and complex in nature. While decentralized and distributed controllers eliminate the disadvantages caused by centralized controller and add plug-and-play functionality, flexibility, and reduction in communication requirements. Each DG has a local controller in the distributed controller (consensus-based control), but the decentralized controller (master-slave control) employs a local controller to control all DGs. The communication links are established across each DG, which results in tracking of reference values of system parameters even in the presence of disturbances. These controllers are connected among themselves through two-way communication networks and exchange local information among themselves. In some cases, the decentralized controller is also impossible to implement because the operation of DGs in the MG is highly dependent on each other and can cause the system to become unstable or operate non-optimally. This makes the distributed controller more dedicated and reliable than the decentralized controller.

It is necessary to control the voltage (which is a local parameter), frequency (global parameter), and power quality (involves proportional sharing of active and reactive power) in the MG. The active power-frequency control and reactive power-voltage droop-based control are used in the primary layer of a hierarchical control scheme. The nominal frequency of the system is achieved by controlling the active power of the system and the nominal voltage is achieved by controlling the reactive power of the system. The robust controller in the primary layer is incorporated with conventional droop controller to improve the voltage regulation and eliminate the disadvantages of traditional controllers, i.e., line impedance dependency, load-dependent deviations, trade-off between the voltage/frequency deviations, accurate power sharing, and poor voltage regulation [

5]. The challenges arising due to the CPL in the MG are addressed in [34] and are handled by a robust droop controller. It adopts a nonlinear disturbance observer to lump the disturbances in the system. In [35], an improved droop-based controller is proposed to achieve proper power sharing by using voltage and frequency droop controllers with optimal tuning of PI controllers. A robust decentralized droop controller with two degrees of freedom is proposed in [36] to mitigate low-frequency oscillations, which is a combination of a conventional droop function and a robust transient droop function. In [37], [38], a type of robust droop-based controller is proposed, which does not require the global information, reduces the computational and communication burden, and handles control issues arising due to the PnP operation, slow dynamic operation, and communication errors. In [20], [39], [40], a type of two-layer hierarchical controller is proposed for the grid-connected MG, where the primary control level consists of a droop controller and the secondary control level restores the voltage with accurate power sharing and handles the uncertainties in the system parameters. In order to control time delay arising due to the limited discontinuous communication among DGs, a two-layer self-consistent control is proposed in [41], which only depends on the local information. In [42], a robust neighbour-based cooperative controller is proposed to deal with communication delay. A resilient centralized robust hierarchical controller is proposed in [43], which works on time-varying network topology in the presence of uncertain communication.

2) Robust Sliding Mode Control

It is a robust design method that can survive external disturbances and parametric uncertainties in both linearized and nonlinearized systems or constrained norms for both deterministic and nondeterministic uncertainties. The implementation of sliding mode control involves selecting a hypersurface or a manifold so that the system trajectory exhibits desirable behavior. The sliding mode control has been intensively investigated for robust nonlinear control to ensure stability subject to parameter constraints. In [

44], [45], a type of robust sliding mode control is proposed in the presence of an unbalanced nonlinear load, as well as the harmonics and negative-sequence current from the unbalanced load. In [44], a sliding mode control is used to control positive-sequence voltage and power, while a combination of Lyapunov function theory and fractional-order sliding mode control is used to control negative-sequence current with harmonics. In [46], [47], a type of decentralized second-order control method based on the sliding mode control is proposed, which is fully distributed in nature, i.e., it does not depend on the global information and it shows better robustness against uncertainties due to the switching function of PWM of VSI. Selecting a proper sliding surface and feedforward controller can control the dynamic behavior with the minimization of chattering.

Figure 7 shows the block diagram of the robust sliding mode controller for restoring the voltage and frequency of the MG [

46]. αv and βv are the control coefficients; pv,qv>0 are the orders of the consensus error; uiv,uiq,uiω,uiSOC,Viref, and ωiref are the voltage, reactive power, frequency, SOC control laws, nominal voltage, and frequency of the ith DG, respectively; vref is the global reference voltage; and Vinom is the control input.

Fig. 7  Block diagram of robust sliding mode controller for restoring voltage and frequency of MG.

3) Robust H Control

The aim of H control is to minimize the impact of uncertainties and the disturbance in the system transient- and steady-state performance. H controllers are based on the synthesis approach for achieving robust performance or stabilization of a multi-variable linear system. In a nonlinear system, the robustness can be obtained in the state space by solving the Riccati equation of LMI technique. In [

48], a robust controller is proposed for the MG against different challenges, i.e., external disturbances and modeling uncertainties. In the case of hybrid AC/DC MG, a multi-variable H control is proposed to adjust the frequency of an islanded MG, which is based on the LMI approach [49], where the uncertainties in SOC are taken into consideration by analysis. A robust H controller is proposed in [50] against uncertainties and unmodelled dynamics. It considers two constraints, i.e., the regional pole placement problem to improve the transient response and the H2 synthesis problem to reduce the control effort against dealing with the uncertainties that track the reference command of DG output voltage, even in the presence of those challenges.

4) Robust Backstepping Control

The backstepping control is based on the recursive Lyapunov method, which allows the system to work in undesirable conditions with flexibility. It is used to extend the kinematic-level controller to dynamic case and to deal with the parameter uncertainties. A robust backstepping controller is proposed in [

51] for the AC/DC MG, which is based on feedback linearization method to maintain power balance and achieve the desired voltage at the PCC. It shows robustness towards parametric uncertainties and external disturbances. A nonlinear integral backstepping controller is proposed for the DC MG in [52], restoring the nominal voltage and handling the power balance under variable power generation simultaneously. A fault-tolerant-based backstepping controller is proposed in [53] to regulate the voltage of the AC MG in the presence of actuator faults and disturbance loads with harmonic/inter-harmonic currents. It does not need accurate modeling of faults, harmonics, or inter-harmonics. So, it improves the reliability of the system. In [54], a backstepping controller is proposed, which considers the uncertainties and disturbances in the disturbance observer, rejects challenges coming from load dynamics, and restores the voltage in the DC MG.

5) Robust Adaptive Control

The adaptation mechanism relies on prior information about the system, whereas the robust adaptive control involves modifying the control law with regard to the change in parameters of the system, unlike conventional one that uses the dynamic gain controller. Though it is less dependent on the mathematical models of the system, it is the most suitable for the MG system encountered with various challenges with unknown state parameters. In [

55], [56], a type of robust adaptive controller is proposed to adjust droop characteristics of the MG. The controller in [55] satisfies both current sharing and bus voltage stability, while the one in [56] regulates the system voltage and frequency to their nominal values after system load variations under various operation conditions. Integration of electric vehicles into existing grid brings challenges to the MG operation. In [57], a robust controller is proposed that considers the effect of electric vehicles and handles load frequency in an islanded MG. The presence of unknown nonlinear functions, unknown control gains, and unknown actuator failures brings instability to the MG operation. A distributed robust adaptive controller is developed in [58] by combining the backstepping method and the dynamic surface control approach in a multi-agent-based MG system. A robust adaptive controller is proposed in [59], which detects mechanical power fluctuations of wind turbines. For optimal functioning, the controller constantly improves the proportional, integral, and derivative gains.

In [

60], an adaptive super-twisting controller is proposed. It takes an extended state observer (ESO) to handle the system uncertainties and measurement noise. Figure 8 shows the block diagram of the ESO-based adaptive super-twisting controller, which restores the nominal voltage and frequency of DGs in the presence of parametric uncertainties and disturbances. In Fig. 8, v^odi and v^odj are the estimated voltages of vodi and vodj, respectively; ei,1 and ei,2 are the cooperating tracking errors between DGs; c,d,m,q>0; si is the sliding surface; vj is the virtual controller parameter; aij and bi are the communication weights and the pinning gains among DGs, respectively; and α and ρi are the controller parameters.

Fig. 8  Block diagram of ESO-based adaptive super-twisting controller [

60].

6) Adaptive Hierarchical Control

The conventional droop control has several disadvantages that are discussed in Section II-B-1). The conventional droop control is modified to the adaptive droop control to improve the voltage and frequency regulation, which is not affected by the system parameter under dynamic perturbation of the MG.

In [

61], an adaptive derivative term of active and reactive power is included to eliminate the considerably large black start transients and circulation currents. To control the power factor at the PCC, an adaptive droop controller is proposed in [62] for grid-connected and islanded MGs.

A type of distributed adaptive voltage controller for the MG is proposed in [

63] and [64], which is fully adaptive and distributed as it does not require the information of DG. It updates the droop coefficient, synchronizes the per-unit current, and provides the proportional load sharing.

Figure 9 shows the block diagram of the adaptive droop controller [

64]. In Fig. 9, v¯odj is the estimated average value of inverter output voltage vodi; vinom is the input to the primary controller; viref is the reference value of DG voltage; δvi is the voltage correction term; vid is the droop voltage; δri is the impedance correction term; ii and ij are the per-unit currents of the ith and jth converters, respectively; and b and aij are the communication weights. The power management and energy balancing issues that arise from coordinating several batteries in an islanded MG are handled by a double-layer hierarchical control technique in [65]. In the primary layer, an adaptive voltage droop is proposed to regulate common bus voltage and maintain SOC level of batteries. The combination of adaptive droop controller with sliding mode control in [66] provides the robustness against voltage fluctuation at DC bus and the fast dynamic response to instantaneously manipulate the output voltage and the input current of each converter.

Fig. 9  Block diagram of adaptive droop controller.

A type of improved mode-adaptive droop controller is proposed in [

67]-[69], which restores the bus voltage regulation and the power sharing without communication infrastructure, maintains the power balance, and accurately regulates the DC bus voltages under various operation conditions.

The adaptive hierarchical control in the secondary layer can provide automated tuning of control gains in real time with regard to the uncertainties/nonlinearities and the external disturbances. In [

70], an adaptive hierarchical control method is proposed by considering generation cost, where the secondary controller is based on the finite-time theory, an improved gray wolf optimization algorithm is used as the tertiary controller to dynamically optimize the rated power of the DG, and the objective function and constraints are established. An adaptive hierarchical energy management strategy for a plug-in heavy electric vehicle is proposed in [71], which is based on the deep learning and genetic algorithm to derive the power split controls between the battery and internal combustion engine. In [72], a robust hierarchical control scheme is proposed for off-grid AC MGs. In the secondary control layer, a robust controller is designed based on an adaptive backstepping integral nonsingular fast-terminal-sliding-mode control strategy to regulate and track the reference of current signals in the presence of unknown bounded uncertainties and external disturbances. An enhanced hierarchical control based energy management scheme is proposed in [73], which is suitable for MG operations during an islanded and grid-connected operation.

7) Adaptive Sliding Mode Control

The robust control design will benefit from the use of adaptive control in terms of performance improvements and extension of the operation range. In [

74], to achieve the rapid and proper dynamic response with regard to external disturbances and uncertainties and the steady-state performance of an unbalanced multi-bus MG, an adaptive sliding mode control is proposed. Similarly, to withstand system uncertainties such as load parameter variations and unmodeled dynamics, an adaptive higher-order sliding mode control is proposed in [75]. The advantage is that the upper bounds of uncertainties are not required during the implementation of the control. An adaptive sliding mode control based on large signal model is proposed in [76] to improve the control dynamics and stability of the DC bus voltage. The steady bus voltage and constant switching frequency are accomplished by integrating sliding mode control with adaptive observation no matter how the parameters vary with the variable CPL and uncertainties. Successful grid synchronization is achieved by an adaptive sliding mode controller in [77] even in the presence of unbalanced loading and grid voltage distortion. In [78], to maintain the active and reactive power regulation and busbar voltage regulation of DGs in grid-connected multi-MG, an adaptive sliding mode controller is considered. An adaptive controller is designed in [79] to observe external disturbances and internal perturbation to enhance the disturbance-rejection performance. To restore the nominal voltage and frequency of the MG in [80], [81], a type of adaptive sliding mode based control is proposed to estimate the reference source current and handles arbitrary communication topology of the MG.

8) Adaptive Backstepping Control

In [

82], [83], a type of adaptive backstepping sliding mode control is proposed to improve the stability from negative impedance characteristics of the bus voltage in the DC MG consisting of CPL. The backstepping controller is implemented for the zero-sequence dynamic stability of the system under different output functions in [84]. An adaptive backstepping controller based on online parameter estimation is proposed in [85], taking into account uncertainties from PnP operation, load fluctuation, and disturbances.

9) Adaptive Event-triggered Control

Though the event-triggered control improves the communication and control efficiencies, in case of disturbances and uncertainties in the system, the triggers run whenever the database fields are updated, which makes the system run slower. However, hybrid adaptive and event-triggered control makes the execution of database update faster by auto-updating and better use of the communication bandwidth. An event-triggered super-twisting sliding mode control is proposed in [

86] to regulate the DC bus voltage and enhance the MG capability to perform stable operation in the presence of matched and unmatched uncertainties in grid-connected and islanded modes. To handle a class of uncertain nonlinear systems, an adaptive event-triggered fault tolerant control is proposed in [87]. It is the combination of event-triggered theory, adaptive backstepping technique, Lyapunov theory, and a novel fault-tolerant control, which together compensate for the effect of actuator faults. In [88]-[90], a type of adaptive event-triggered communication-based distributed secondary cooperative control is presented to handle a variety of objectives, including estimating the states to reduce communication burden [88], managing economic dispatch [89], and dealing with periodic disturbances [90]. A dynamic event-triggered communication scheme provides a good balance between the dynamic performance and communication burden in [91]. A type of robust sliding mode control based on the adaptive event-triggered mechanism is proposed in [92], [93] against the frequency deviation caused by power unbalance or time delays in a multi-area interconnected MG.

10) Adaptive Passivity-based Control

The passivity-based control is favored because of its low complexity. But it cannot eliminate the steady-state error of the output voltage caused by the system disturbances such as load and line variations. The conventional passivity-based control strategies, which are combined with the integral gain control, attenuate the system disturbance through the feedback regulation, which is relatively slow. In [

94], an adaptive passivity-based control is designed, where a nonlinear disturbance observer improves the control robustness against both load and line variations to achieve system stability. It ensures large signal stability as well as fast recovery performance of the system during disturbance/uncertainty as compared to other nonlinear control methods. To mitigate the instability effect of power converter supplying a CPL and a constant voltage load in the DC MG, a type of adaptive passivity-based control is proposed in [95], [96]. In [97], a nonlinear adaptive controller is proposed based on passivity-based control. The system injects the predetermined values of active/reactive power from a DC grid to the AC grid while the voltage of the interlinking capacitor is fixed.

B. Remarks

The robust and adaptive controllers exhibit robustness in the presence of uncertainties and disturbances. The robust droop controller eliminates the disadvantages of conventional droop controllers, mitigates the lower frequency oscillations [

36], and does not depend upon the global information of the MG system [38]. Whereas the adaptive droop controller adds several benefits such as fast dynamic response [56] and performance remaining unaffected by system impedance [62]. The robust backstepping controller acts as an independent nonlinear controller that is independent of accurate modeling of faults and harmonics [46], whereas the adaptive backstepping control can identify the system parameters and is able to track the response of those parameters despite unbounded uncertainties [78]. The distributed sliding mode control not only improves the chattering as compared to the conventional sliding mode control but also shows better robustness towards uncertainties, i.e., noise and time delay [45], [46]. Whereas the combination of the adaptive control and sliding mode control improves the steady-state performance [69], and achieves fast grid synchronization as compared to the adaptive control [71].

The robustness of MG can also be improved by employing artificial intelligence techniques like neural network, fuzzy logic, and machine learning. In [

110], a reinforcement learning approach is proposed to coordinate current sharing and voltage restoration in an islanded DC MG. It is an adaptive learning approach to handle the disturbance of the system, which does not depend on mathematical modeling. An adaptive neuro-fuzzy inference system is implemented in [111] to effectively control the frequency and power sharing among DG units. In [112], an adaptive fuzzy controller is proposed for the coordinated power flow control among the generation, demand, and storage system. It also restores the load frequency and MG frequency in the presence of disturbance and parametric uncertainties. In [113], an adaptive neural network controller is proposed for tracking the maximum power point of renewable energy generators and the optimal power exchange between converters and grid, regulating DC-link output voltage, and mitigating voltage and power oscillations after disturbances, in which the stability is achieved by the local measurement converter. Although it shows robustness towards system challenges with less computation, the intelligent-based control is still robust but not adaptive. Besides, the real-time implementation of neural network is impossible, and for the multi-MG system, the controller implementation becomes complex as compared to the robust and adaptive control. In addition to that, the implementation of above control schemes in a practical setting offers significant challenges in the MG such as the effects of line impedances and electromagnetic field, considering the trade-off of system parameters, and the under- or over-saturation of sensors and actuators because of the high gain of controllers and the ageing of sensors and actuators.

V. Future Trends in Robust and Adaptive Control

A brief summary of future trends of robust and adaptive controllers in the field of MG control is discussed below.

1) Non-uniform communication delay and mixed communication constraints. The time delay is unknown and encountered in different approaches, so the control should provide enough robustness towards the non-uniform communication delay and random delays, where the upper and lower bounds are uncertain [

3], [8]. Communication delay, disturbances, and uncertainties are very important in analyzing system parameters of the MG. Hence, mixed communication constraints need to be addressed while modeling the MG, and various stability analyses also need to be explored for such a scenario including delay and loss rate.

2) Fast switching topologies and limited communication bandwidth. Various types of switching operations are needed for successful operations of the MG. The PnP functionality of DERs can ensure the stability of the MG system [

3], which should be done so that it will not affect the existing MG. The fast switching often leads to a large state transient. Besides, the presence of limited communication bandwidth leads to the communication congestion and increased packet loss [48]. Therefore, the MG encountered with such challenges is an open field for future research.

3) Automation control against cyber attacks. There are various types of cyber attacks in the cyber-enabled MG. Among them, the denial-of-service and false data injection attacks are more common [

11]. Several research works are reported to handle such type of cyber attacks in the MG. A replay attack is a type of attack where the attackers deliberately delay or repeat the data transmission. There is less literature available that addresses the issue on cyber attacks of the MG.

4) Scalability. The scalability of a controller, i.e., how the system behaves when the complexity and data burden increase, has not been addressed. Thus, it is important to consider the multi-MG. With the shifting of research to model-free control, data-driven and machine learning approaches play a significant role in the control and management of the MG [

110]. In order to reduce the communication burden, decrease the communication cost, and increase the delay margin, the robust controller with the machine learning approach or adaptive intelligent approach is the future trend.

Table II presents the current literature in the field of MG against the communication constraints, uncertainties, and disturbances, which provides an overall picture of the control challenges in designing control systems.

Table II  Current Literature in Field of MG Against Communication Constraints, Uncertainties, and Disturbances
ChallengeKey featureMG objectiveControl schemeReference
Uncertain communication links AC, islanded MG Frequency Adaptive control [98]
Aperiodic sampled time delay AC, islanded MG Voltage and frequency Robust cooperative control [99]
Multiple communication delays and uncertainties AC, islanded MG Voltage Robust integral sliding mode control [26]
Communication delay AC, islanded MG Frequency and load sharing Distributed cooperative control [100]
Time-varying delays, data packet losses, and link failures AC, islanded MG Voltage, frequency, and active power sharing Model predictive control [101]
Sensor fault, data attack AC, islanded MG Voltage and frequency Adaptive resilient control [29]
Data disturbance and time delay AC, islanded MG Voltage and frequency Robust H control [27]
Cyber attacks AC, islanded MG Frequency and load sharing Consensus based robust control [102]
Parametric uncertainty and exogenous disturbances DC, islanded MG Voltage Robust centralized control [103]
Communication and system noise AC, islanded MG Voltage and frequency Robust cooperative control [104]
External disturbances AC, islanded MG Voltage and frequency Robust fractional sliding control [105]
External disturbance, uncertainty, and fixed time delays AC, islanded MG Voltage, frequency, and SOC Robust H control [28]
Parameter uncertainties and external disturbances DC, islanded MG Voltage Robust voltage stabilization control [106]
Uncertainties and disturbances AC, islanded MG Voltage, frequency, and active power sharing Distributed adaptive secondary control [107]
Topological, operational uncertainties, and latency AC, grid-connected MG Load current Decentralized adaptive control [30]
Malicious time delays and packet loss AC, islanded MG Voltage and frequency Robust data prediction control [108]
PnP AC, islanded MG Voltage Extended proportional-derivative control [109]
Disturbances, uncertainties, noises, delays, and attacks AC, islanded MG Voltage, frequency, active and reactive power Resilient adaptive consensus control [31]

VI. Conclusion

This paper presents a comprehensive review of different challenges encountered during the integration of RESs to the existing grid with regard to the communication constraints during the MG operation and control. Various robust and adaptive control schemes to handle the control issues are presented, as well as the current research trends in the field of robust and adaptive control. It can be concluded that to achieve control objectives of the MG and overcome the above challenges, robust and adaptive controllers show significantly improved performance in terms of transient and steady-state behavior and robustness as compared to traditional controllers.

References

1

V. A. Suryad, S. Doolla, and M. Chandorkar, “Microgrids in India: possibilities and challenges,” IEEE Electrification Magazine, vol. 5, no. 2, pp. 47-55, Jun. 2017. [Baidu Scholar] 

2

R. Badal, P. Das, S. K. Sarker et al., “A survey on control issues in renewable energy integration and microgrid,” Protection and Control of Modern Power Systems, vol. 4, no. 1, pp. 87-113, Dec. 2019. [Baidu Scholar] 

3

X. Lu and J. Lai, “Communication constraints for distributed secondary control of heterogenous microgrids: a survey,” IEEE Transactions on Industry Applications, vol. 57, no. 6, pp. 5636-5648, Aug. 2021. [Baidu Scholar] 

4

F. Gao, R. Kang, J. Cao et al., “Primary and secondary control in DC microgrids: a review,” Journal of Modern Power Systems and Clean Energy, vol. 7, no. 2, pp. 227-242, Mar. 2019. [Baidu Scholar] 

5

S. Shrivastava and B. Subudhi, “Comprehensive review on hierarchical control of cyber-physical microgrid system,” IET Generation, Transmission & Distribution, vol. 14, no. 26, pp. 6397-6416, Nov. 2020. [Baidu Scholar] 

6

T. K. Roy, M. A. Mahmud, A. M. Oo et al., “Nonlinear adaptive backstepping controller design for controlling bidirectional power flow of BESSs in DC microgrids,” in Proceedings of IEEE Industry Applications Society Annual Meeting, Portland, USA, Oct. 2016, pp. 1-8. [Baidu Scholar] 

7

M. Huang and L. Ding, “Distributed observer-based fault-tolerant control for DC microgrids with sensor fault,” IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 68, no. 4, pp. 1659-1670, Jan. 2021. [Baidu Scholar] 

8

A. Afshari, M. Karrari, H. R. Baghaee et al., “Distributed fault-tolerant voltage/frequency synchronization in autonomous AC microgrids,” IEEE Transactions on Power Systems, vol. 35, no. 5, pp. 3774-3789, Sept. 2020. [Baidu Scholar] 

9

M. A. Shahab, B. Mozafari, S. Soleymani et al., “Distributed consensus-based fault tolerant control of islanded microgrids,” IEEE Transactions on Smart Grid, vol. 11, no. 1, pp. 37-47, Jan. 2020 [Baidu Scholar] 

10

S. Zuo, T. Altun, F. L. Lewis et al., “Distributed resilient secondary control of DC microgrids against unbounded attacks,” IEEE Transactions on Smart Grid, vol. 11, no. 5, pp. 3850-3859, May 2020. [Baidu Scholar] 

11

H. F. Habib, C. R. Lashway, and O. A. Mohammed, “A review of communication failure impacts on adaptive microgrid protection schemes and the use of energy storage as a contingency,” IEEE Transactions on Industry Applications, vol. 54, no. 2, pp. 1194-1207, Nov. 2017. [Baidu Scholar] 

12

N. M. Dehkordi, H. R. Baghaee, N. Sadati et al., “Distributed noise-resilient secondary voltage and frequency control for islanded microgrids,” IEEE Transactions on Smart Grid, vol. 10, no. 4, pp. 3780- 3790, Jul. 2019. [Baidu Scholar] 

13

J. Lai and X. Lu, “Communication constraints for distributed secondary control of heterogenous microgrids: a brief survey,” in Proceedings of IEEE IAS 57th Industrial and Commercial Power Systems Technical Conference, Las Vegas, USA, Apr. 2021, pp. 1-9. [Baidu Scholar] 

14

A. Afshari, M. Karrari, H. R. Baghaee et al., “Resilient cooperative control of AC microgrids considering relative statedependent noises and communication time-delays,” IET Renewable Power Generation, vol. 14, no. 8, pp. 1321-1331, Jun. 2020. [Baidu Scholar] 

15

W. Yao, Y. Wang, Y. Xu et al., “Distributed weight-average-prediction control and stability analysis for an islanded microgrid with communication time delay,” IEEE Transactions on Power Systems, vol. 37, no. 1, pp. 330-342, Jun. 2021. [Baidu Scholar] 

16

A. B. Shyam, S. Anand, and S. R. Sahoo, “Effect of communication delay on consensus-based secondary controllers in DC microgrid,” IEEE Transactions on Industrial Electronics, vol. 68, no.4, pp. 3202-3212, Mar. 2020. [Baidu Scholar] 

17

J. Lai, X. Lu, X. Yu et al., “Distributed voltage regulation for cyber-physical microgrids with coupling delays and slow switching topologies,” IEEE Transactions on Systems, Man, and Cybernetics: Systems, vol. 50, no. 1, pp. 100-110, Jan. 2020. [Baidu Scholar] 

18

Y. Cheng, R. Azizipanah-Abarghooee, S. Azizi et al., “Smart frequency control in low inertia energy systems based on frequency response techniques: a review,” Applied Energy, vol. 279, p. 115798, Dec. 2020. [Baidu Scholar] 

19

Q. Zhou, M. Shahidehpour, A. Paaso et al., “Distributed control and communication strategies in networked microgrids,” IEEE Communications Surveys and Tutorials, vol. 22, no. 4, pp. 2586-2633, Sept. 2020. [Baidu Scholar] 

20

Y. A.-R. I. Mohamed and A. A. Radwan, “Hierarchical control system for robust microgrid operation and seamless mode transfer in active distribution systems,” IEEE Transactions on Smart Grid, vol. 2, no. 2, pp. 352-362, May 2011. [Baidu Scholar] 

21

F. Nejabatkhah, Y. Li, H. Liang et al., “Cyber-security of smart microgrids: a survey,” Energies, vol. 14, no. 1, pp. 1-27, Dec. 2021. [Baidu Scholar] 

22

X. Liu, M. Shahidehpour, Y. Cao et al., “Microgrid risk analysis considering the impact of cyber attacks on solar PV and ESS control systems,” IEEE Transactions on Smart Grid, vol. 8, no. 3, pp. 1330-1339, Oct. 2017. [Baidu Scholar] 

23

C. Natesan, S. Ajithan, and P. Palani, “Survey on microgrid: power quality improvement techniques” ISRN Renewable Energy, vol. 2014, pp. 1-7, Jan. 2014. [Baidu Scholar] 

24

B. Adineh, R. Keypour, P. Davari et al., “Review of harmonic mitigation methods in microgrid: from a hierarchical control perspective,” IEEE Journal of Emerging and Selected Topics in Power Electronics, vol. 9, no. 3, pp. 3044-3060, Jun. 2020. [Baidu Scholar] 

25

F. Mohammadi, B. Mohammadi-Ivatloo, G. B. Gharehpetian et al., “Robust control strategies for microgrids: a review,” IEEE Systems Journal, vol. 16, no. 2, pp. 2401-2412, Jun. 2021. [Baidu Scholar] 

26

M. Gholami, A. Pilloni, A. Pisano et al., “Robust distributed secondary voltage restoration control of AC microgrids under multiple communication delays,” Energies, vol. 14, p. 1165, Jan. 2021. [Baidu Scholar] 

27

G. Cao, G. Lou, W. Gu et al., “H robustness for distributed control in autonomous microgrids considering cyber disturbances,” CSEE Journal of Power and Energy Systems. doi: 10.17775/CSEEJPES.2020.04110 [Baidu Scholar] 

28

M. Raeispour, H. Atrianfar, H. R. Baghaee et al., “Resilient-consensus-based control of autonomous AC microgrids with uncertain time-delayed communications,” IEEE Transactions on Smart Grid, vol. 11, no. 5, pp. 3871-3884, Mar. 2020. [Baidu Scholar] 

29

X. Li, Q. Xu, and F. Blaabjerg, “Adaptive resilient secondary control for islanded AC microgrids with sensor faults,” IEEE Journal of Emerging and Selected Topics in Power Electronics, vol. 9, no. 5, pp. 5239-5248, Apr. 2020. [Baidu Scholar] 

30

S. A. Hosseini, S. H. H. Sadeghi, and A. Nasiri, “Decentralized adaptive protection coordination based on agents social activities for microgrids with topological and operational uncertainties,” IEEE Transactions on Industry Applications, vol. 57, no. 1, pp. 702-713, Oct. 2020. [Baidu Scholar] 

31

N. M. Dehkordi and S. Z. Moussavi, “Distributed resilient adaptive control of islanded microgrids under sensor/actuator faults,” IEEE Transactions on Smart Grid, vol. 11, no. 3, pp. 2699-2708, Dec. 2019. [Baidu Scholar] 

32

P. Dorato, “A historical review of robust control,” IEEE Control Systems, vol. 7, no. 2, pp. 44-47, Apr. 1987. [Baidu Scholar] 

33

S. G. Anavatti, F. Santoso, and M. A. Garratt, “Progress in adaptive control systems: past, present, and future,” in Proceedings of IEEE International Conference on Advanced Mechatronics, Intelligent Manufacture, and Industrial Automation (ICAMIMIA), Surabaya, Indonesia, Oct. 2015, pp. 1-8. [Baidu Scholar] 

34

Q. Xu, C. Zhang, and P. Wang, “A robust droop-based autonomous controller for decentralized power sharing in DC microgrid considering large-signal stability,” IEEE Transactions on Industrial Informatics, vol. 16, no. 3, pp. 1483-1494, Oct. 2019. [Baidu Scholar] 

35

F. Mohammadi, G. A. Nazri, and M. Saif, “An improved droop-based control strategy for MTHVDC systems,” Electronics, vol. 9, no. 1, p. 87, Jan. 2020. [Baidu Scholar] 

36

N. M. Dehkordi, N. Sadati, and M. Hamzeh, “Robust tuning of transient droop gains based on Kharitonov’s stability theorem in droop-controlled microgrids,” IET Generation, Transmission & Distribution, vol. 12, no. 14, pp. 3495-3501, Aug. 2018. [Baidu Scholar] 

37

J. Hu and A. Lanzon, “Distributed finite-time consensus control for heterogeneous battery energy storage systems in droop-controlled microgrids,” IEEE Transactions on Smart Grid, vol. 10, no. 5, pp. 4751-4761, Sept. 2019. [Baidu Scholar] 

38

Q. Xu, Y. Xu, C. Zhang et al., “A robust droop-based autonomous controller for decentralized power sharing in DC microgrid considering large-signal stability,” IEEE Transactions on Industrial Informatics, vol. 16, no. 3, pp. 1483-1494, Mar. 2020. [Baidu Scholar] 

39

H. Cai and G. Hu, “Distributed robust hierarchical power sharing control of grid-connected spatially concentrated AC microgrid,” IEEE Transactions on Control Systems Technology, vol. 27, no. 3, pp. 1012-1022, May 2019. [Baidu Scholar] 

40

Y. Xu, Q. Guo, H. Sun et al., “Distributed discrete robust secondary cooperative control for islanded microgrids,” IEEE Transactions on Smart Grid, vol. 10, no. 4, pp. 3620-3629, Jul. 2019. [Baidu Scholar] 

41

J. Lai, X. Lu, X. Yu et al., “Distributed voltage regulation for cyber-physical microgrids with coupling delays and slow switching topologies,” IEEE Transactions on Systems, Man, and Cybernetics: Systems, vol. 50, no. 1, pp. 100-110, Jan. 2020. [Baidu Scholar] 

42

M. Raeispour, H. Atrianfar, H. R. Baghaee et al., “Robust hierarchical control of VSC-based off-grid AC microgrids to enhancing stability and FRT capability considering time-varying delays,” IEEE Journal of Emerging and Selected Topics in Power Electronics, vol. 9, no. 6, pp. 7159-7172, Aug. 2020. [Baidu Scholar] 

43

M. Raeispour, H. Atrianfar, H. R. Baghaee et al., “Robust distributed disturbance-resilient-based control of off-grid microgrids with uncertain communications,” IEEE Systems Journal, vol. 5, no. 2, pp. 2895-905, Mar. 2021. [Baidu Scholar] 

44

M. Cucuzzella, G. P. Incremona, and A. Ferrara, “Decentralized sliding mode control of islanded AC microgrids with arbitrary topology,” IEEE Transactions on Industrial Electronics, vol. 64, no. 8, pp. 6706-6713, Aug. 2017. [Baidu Scholar] 

45

H. R. Baghaee, M. Mirsalim, G. B. Gharehpetian et al., “Decentralized sliding mode control of WG/PV/FC microgrids under unbalanced and nonlinear load conditions for on- and off-grid modes,” IEEE Systems Journal, vol. 12, no. 4, pp. 3108-3119, Dec. 2018. [Baidu Scholar] 

46

R. Zhang and B. Hredzak, “Nonlinear sliding mode and distributed control of battery energy storage and photovoltaic systems in AC microgrids with time delays,” IEEE Transactions on Industrial Informatics, vol. 15, no. 9, pp. 5149-5160, Sept. 2019. [Baidu Scholar] 

47

A. E. M. Bouzid, P. Sicard, H. Chaoui et al., “Robust three degrees of freedom based on controller of voltage/current loops for DG unit in micro grids,” IET Power Electronics, vol. 12, no. 6, pp. 1413-1424, May 2019. [Baidu Scholar] 

48

B. E. Sedhom, M. M. El-Saadawi, A. Y. Hatata et al., “Robust control technique in an autonomous microgrid: a multi-stage controller based on harmony search algorithm,” Iranian Journal of Science and Technology, Transactions of Electrical Engineering, vol. 44, no. 1, pp. 377-402, Mar. 2020. [Baidu Scholar] 

49

Q. L. Lam, A. I. Bratcu, D. Riu et al., “Primary frequency control in stand-alone micro grids with storage units: a robustness analysis confirmed by real-time experiments,” International Journal of Electrical Power & Energy Systems, vol. 115, p. 105507, Feb. 2020. [Baidu Scholar] 

50

M. Raeispour, H. Atrianfar, H. R. Baghaee et al., “Robust sliding mode and mixed output feedback primary control of AC microgrids,” IEEE Systems Journal, vol. 15, no. 2, pp. 1-12, Jun. 2020. [Baidu Scholar] 

51

N. M. Dehkordi, N. Sadati, and M. Hamzeh, “Robust backstepping control of an interlink converter in a hybrid AC/DC microgrid based on feedback linearisation method,” International Journal of Control, vol. 90, no. 9, pp. 1990-2004, Sept. 2017. [Baidu Scholar] 

52

H. Armghan, M. Yang, M. Wang et al., “Nonlinear integral backstepping based control of a DC microgrid with renewable generation and energy storage systems,” International Journal of Electrical Power & Energy Systems, vol. 117, p. 105613, May 2020. [Baidu Scholar] 

53

N. M. Dehkordi and V. Nekoukar, “Robust reliable fault tolerant control of islanded microgrids using augmented backstepping control,” IET Generation, Transmission & Distribution, vol. 14, no. 3, pp. 432-440, Feb. 2020. [Baidu Scholar] 

54

H. Amiri, G. A. Markadeh, N. M. Dehkordi et al., “Fully decentralized robust backstepping voltage control of photovoltaic systems for DC islanded microgrids based on disturbance observer method,” ISA Transaction, vol. 101, pp. 471-481, Jun. 2020. [Baidu Scholar] 

55

T. V. Vu, D. Perkins, F. Diaz et al., “Robust adaptive droop control for DC microgrids,” Electric Power Systems Research, vol. 146, pp. 95-106, May 2017. [Baidu Scholar] 

56

B. E. Sedhom, A. Y. Hatata, M. M. El‐Saadawi et al., “Robust adaptive H-infinity based controller for islanded microgrid supplying non-linear and unbalanced loads,” IET Smart Grid, vol. 2, no.3, pp. 420-435, Oct. 2019. [Baidu Scholar] 

57

M. H. Khooban, T. Niknam, F. Blaabjerg et al., “A robust adaptive load frequency control for micro-grids,” ISA Transaction, vol. 65, pp. 220-229, Nov. 2016. [Baidu Scholar] 

58

M. Hashemi and G. Shahgholian, “Distributed robust adaptive control of high order nonlinear multi agent systems,” ISA Transaction, vol. 74, pp. 14-27, Mar. 2018. [Baidu Scholar] 

59

Z. A. Alrowaili, M. M. Ali, A. Youssef et al., “Robust adaptive HCS MPPT algorithm-based wind generation system using model reference adaptive control,” Sensors, vol. 21, no. 15, p. 5187, Jan. 2021. [Baidu Scholar] 

60

P. Ge, X. Dou, X. Quan et al., “Extended-state-observer-based distributed robust secondary voltage and frequency control for an autonomous microgrid,” IEEE Transactions on Sustainable Energy, vol. 11, no. 1, pp. 195-205, Dec. 2020. [Baidu Scholar] 

61

Y. A. R. I. Mohamed and E. F. El-Saadany, “Adaptive decentralized droop controller to preserve power sharing stability of paralleled inverters in distributed generation microgrids,” IEEE Transactions on Power Electronics, vol. 23, no. 6, pp. 2806-2816, Dec. 2008. [Baidu Scholar] 

62

J. Kim, J. M. Guerrero, P. Rodriguez et al., “Mode adaptive droop control with virtual output impedances for an inverter-based flexible AC microgrid,” IEEE Transactions on Power Electronics, vol. 26, no. 3, pp. 689-701, Nov. 2010. [Baidu Scholar] 

63

A. Bidram, A. Davoudi, F. L. Lewis et al., “Distributed adaptive voltage control of inverter-based microgrids,” IEEE Transactions on Energy Conversion, vol. 29, no. 4, pp. 862-872, Oct. 2014. [Baidu Scholar] 

64

V. Nasirian, A. Davoudi, F. L. Lewis et al., “Distributed adaptive droop control for DC distribution systems,” IEEE Transactions on Energy Conversion, vol. 29, no. 4, pp. 944-956, Sept. 2014. [Baidu Scholar] 

65

S. Augustine, M. K. Mishra, and N. Lakshminarasamma, “Adaptive droop control strategy for load sharing and circulating current minimization in low-voltage standalone DC microgrid,” IEEE Transactions on Sustainable Energy, vol. 6, no. 1, pp. 132-141, Nov. 2014. [Baidu Scholar] 

66

M. Mokhtar, M. I. Marei, and A. A. El-Sattar, “An adaptive droop control scheme for DC microgrids integrating sliding mode voltage and current controlled boost converters,” IEEE Transactions on Smart Grid, vol. 10, no. 2, pp. 1685-1693, Nov. 2017. [Baidu Scholar] 

67

J. Mohammadi and F. B. Ajaei, “Improved mode-adaptive droop control strategy for the DC microgrid,” IEEE Access, vol. 7, pp. 86421-86435, Jun. 2019. [Baidu Scholar] 

68

J. C. Vasquez, J. M. Guerrero, A. Luna et al., “Adaptive droop control applied to voltage-source inverters operating in grid-connected and islanded modes,” IEEE Transactions on Industrial Electronics, vol. 56, no. 10, pp. 4088-4096, Jul. 2009. [Baidu Scholar] 

69

Y. A.-R. I. Mohamed and E. F. El-Saadany, “Adaptive decentralized droop controller to preserve power sharing stability of paralleled inverters in distributed generation microgrids,” IEEE Transactions on Power Electronics, vol. 23, no. 6, pp. 2806-2816, Dec. 2008. [Baidu Scholar] 

70

L. Ma and J. Zhang, “An adaptive hierarchical control method for microgrid considering generation cost,” IEEE Access, vol. 8, pp. 164187-164199, Sept. 2020. [Baidu Scholar] 

71

T. Liu, X. Tang, H. Wang et al., “Adaptive hierarchical energy management design for a plug-in hybrid electric vehicle,” IEEE Transactions on Vehicular Technology, vol. 68, no. 12, pp. 11513-11522, Jul. 2019. [Baidu Scholar] 

72

M. Raeispour, H. Atrianfar, H. R. Baghaee et al., “Robust hierarchical control of VSC-based off-grid AC microgrids to enhancing stability and FRT capability considering time-varying delays,” IEEE Journal of Emerging and Selected Topics in Power Electronics, vol. 9, no. 6, pp. 7159-7172, Aug. 2020. [Baidu Scholar] 

73

M. Mu, Y. Guan, Y. Terriche et al., “Adaptive power management of hierarchical controlled hybrid shipboard microgrids,” IEEE Access, vol. 10, pp. 21397-21411, Feb. 2022. [Baidu Scholar] 

74

A. H. Eshghi, J. Soltani, M. M. Rezaei et al., “A robust control strategy for a single-phase grid-connected multibus microgrid based on adaptive sliding mode control and dynamic phasor concept,” International Transactions on Electrical Energy Systems, vol. 31, no. 8, p. 12936, Aug. 2021. [Baidu Scholar] 

75

Y. Han, R. Ma, and J. Cui, “Adaptive higher-order sliding mode control for islanding and grid-connected operation of a microgrid,” Energies, vol. 11, no. 6, p. 1459, Jun. 2018. [Baidu Scholar] 

76

D. Zhang and J. Wang, “Adaptive sliding-mode control in bus voltage for an islanded DC microgrid,” Mathematical Problems in Engineering, vol. 2017, no. 8962086, pp. 1-9, Dec. 2017. [Baidu Scholar] 

77

A. Bag, B. Subudhi, and P. K. Ray, “An adaptive sliding mode control scheme for grid integration of a PV system,” CPSS Transactions on Power Electronics and Applications, vol. 3, no. 4, pp. 362-371, Dec. 2018. [Baidu Scholar] 

78

Z. F. Shavakhi, E. Rokrok, J. Soltani et al., “Adaptive sliding mode control of multi-DG, multi-bus grid-connected microgrid,” Journal of Operation and Automation in Power Engineering, vol. 7, pp. 65-77, May 2019. [Baidu Scholar] 

79

Z. Chen, A. Luo, H. Wang et al., “Adaptive sliding-mode voltage control for inverter operating in islanded mode in microgrid,” International Journal of Electrical Power & Energy Systems, vol. 66, pp. 133-143, Mar. 2015. [Baidu Scholar] 

80

U. K. Kalla, B. Singh B, S. S. Murthy et al., “Adaptive sliding mode control of standalone single-phase microgrid using hydro, wind, and solar PV array-based generation,” IEEE Transactions on Smart Grid, vol. 9, no. 6, pp. 6806-6814, Jul. 2017. [Baidu Scholar] 

81

M. Cucuzzella, G. P. Incremona, and A. Ferrara, “Decentralized sliding mode control of islanded AC microgrids with arbitrary topology,” IEEE Transactions on Industrial Electronics, vol. 64, no. 8, pp. 6706-6713, Aug. 2017. [Baidu Scholar] 

82

J. Wu and Y. Lu, “Adaptive backstepping sliding mode control for boost converter with constant power load,” IEEE Access, vol. 7, pp. 50797-50807, Apr. 2019. [Baidu Scholar] 

83

T. K. Roy, M. A. Mahmud, A. M. Oo et al., “Nonlinear adaptive backstepping controller design for islanded DC microgrids,” IEEE Transactions on Industry Applications, vol. 54, no. 3, pp. 2857-2873, Feb. 2018. [Baidu Scholar] 

84

C. Guo, A. Zhang, H. Zhang et al., “Adaptive backstepping control with online parameter estimator for a plug-and-play parallel converter system in a power switcher,” Energies, vol. 11, no. 12, p. 3528, Dec. 2018. [Baidu Scholar] 

85

J. Mao, X. Zhang, T. Dai et al., “An adaptive backstepping sliding mode cascade-control method for a DC microgrid based on nonlinear virtual inertia,” Electronics, vol. 10, no. 24, p. 3100, Jan. 2021. [Baidu Scholar] 

86

V. Kumar, S. R. Mohanty, and S. Kumar, “Event trigger super twisting sliding mode control for DC micro grid with matched/unmatched disturbance observer,” IEEE Transactions on Smart Grid, vol. 11, no. 5, pp. 3837-3849, Apr. 2020. [Baidu Scholar] 

87

C. Zhu, C. Li, X. Chen et al., “Event-triggered adaptive fault tolerant control for a class of uncertain nonlinear systems,” Entropy, vol. 22, no. 6, p. 598, May 2020. [Baidu Scholar] 

88

J. Zhou, Y. Xu, H. Sun et al., “Distributed event-triggered consensus based current sharing control of DC microgrids considering uncertainties,” IEEE Transactions on Industrial Informatics, vol. 16, no. 12, pp. 7413-7425, Dec. 2019. [Baidu Scholar] 

89

S. Sahoo and S. Mishra, “An adaptive event-triggered communication-based distributed secondary control for DC microgrids,” IEEE Transactions on Smart Grid, vol. 9, no. 6, pp. 6674-6683, Jun. 2017. [Baidu Scholar] 

90

Y. Wan, C. Long, R. Deng et al., “Adaptive event-triggered strategy for economic dispatch in uncertain communication networks,” IEEE Transactions on Control of Network Systems, vol. 8, no. 4, pp. 1881-1891, Jun. 2021. [Baidu Scholar] 

91

H. Ma, H. Li, R. Lu et al., “Adaptive event-triggered control for a class of nonlinear systems with periodic disturbances,” Science China Information Sciences, vol. 63, no. 5, pp. 1-15, May 2020. [Baidu Scholar] 

92

C. Yang, W. Yao, J. Fang et al., “Dynamic event-triggered robust secondary frequency control for islanded AC microgrid,” Applied Energy, vol. 242, pp. 821-836, May 2019. [Baidu Scholar] 

93

H. Li, X. Wang, and J. Xiao, “Adaptive event-triggered load frequency control for interconnected microgrids by observer-based sliding mode control,” IEEE Access, vol. 7, pp. 68271-68280, Jun. 2019. [Baidu Scholar] 

94

M. A. Hassan, E. Li, X. Li et al., “Adaptive passivity-based control of DC-DC buck power converter with constant power load in DC microgrid systems,” IEEE Journal of Emerging and Selected Topics in Power Electronics, vol. 7, no. 3, pp. 2029-2040, Oct. 2018. [Baidu Scholar] 

95

M. A. Hassan, C. Su, F. Chen et al., “Adaptive passivity-based control of DC-DC boost power converter supplying constant power and constant voltage loads,” IEEE Transactions on Industrial Electronics, vol. 69, no. 6, pp. 6204-6214, Jun. 2021. [Baidu Scholar] 

96

C. A. Soriano-Rangel, W. He, F. Mancilla-David et al., “Voltage regulation in buck-boost converters feeding an unknown constant power load: an adaptive passivity-based control,” IEEE Transactions on Control Systems Technology, vol. 29, no. 1, pp. 395-402, Jan. 2020. [Baidu Scholar] 

97

S. M. Azimi and M. Hamzeh, “Adaptive interconnection and damping assignment passivity-based control of interlinking converter in hybrid AC/DC grids,” IEEE Systems Journal, vol. 14, no. 4, pp. 4718-4725, Jan. 2020. [Baidu Scholar] 

98

Z. Li and G. Chen, “Distributed adaptive control scheme for islanded AC microgrids with tolerance to uncertain communication links,” IEEE Systems Journal, vol. 16, no. 2, pp. 2741-2752, Dec. 2021. [Baidu Scholar] 

99

R. Zhang and B. Hredzak, “Distributed control system with aperiodic time-delayed sampled data for batteries and renewable energy sources in microgrids,” IEEE Transactions on Sustainable Energy, vol. 11, no. 2, pp. 1013-1022, Apr. 2020. [Baidu Scholar] 

100

J. Lai, X. Lu, and X. Yu, “Stochastic distributed frequency and load sharing control for microgrids with communication delays,” IEEE Systems Journal, vol. 13, no. 4, pp. 4269-4280, Mar. 2019. [Baidu Scholar] 

101

M. Mottaghizadeh, F. Aminifar, T. Amraee et al., “Distributed robust secondary control of islanded microgrids: voltage, frequency, and power sharing,” IEEE Transactions on Power Delivery, vol. 36, no. 4, pp. 2501-2509, Apr. 2021. [Baidu Scholar] 

102

Y. Liu, Y. Li, Y. Wang et al., “Robust and resilient distributed optimal frequency control for microgrids against cyber attacks,” IEEE Transactions on Industrial Informatics, vol. 18, no. 1, pp. 375-386, Apr. 2021. [Baidu Scholar] 

103

M. Mehdi, S. Z. Jamali, M. O. Khan et al., “Robust control of a DC microgrid under parametric uncertainty and disturbances,” Electric Power Systems Research, vol. 179, p. 106074, Feb. 2020. [Baidu Scholar] 

104

N. M. Dehkordi and V. Nekoukar, “Robust distributed stochastic secondary control of microgrids with system and communication noises,” IET Generation, Transmission & Distribution, vol. 14, no. 6, pp. 1148-1158, Mar. 2020. [Baidu Scholar] 

105

H. Delavari and S. Naderian, “Design and HIL implementation of a new robust fractional sliding mode control of microgrids,” IET Generation, Transmission & Distribution, vol. 14, no. 26, pp. 6690-6702, Jan. 2021. [Baidu Scholar] 

106

J. Lee, H.-S. Ahn, and J. Back, “Robust voltage stabilization controller for uncertain DC microgrids,” IEEE Access, vol. 9, pp. 99606-99616, Jul. 2021. [Baidu Scholar] 

107

M. Keshavarz, A. Doroudi, M. H. Kazemi et al., “A novel adaptive distributed secondary voltage controller with high convergence rate for islanded microgrids,” IEEE Systems Journal, vol. 15, no. 3, pp. 4157-67, Oct. 2020. [Baidu Scholar] 

108

R. Heydari, Y. Khayat, A. Amiri et al., “Robust high-rate secondary control of microgrids with mitigation of communication impairments,” IEEE Transactions on Power Electronics, vol. 35, no. 11, pp. 12486-12496, Nov. 2020. [Baidu Scholar] 

109

A. Zarei, Y. Mousavi, R. Mosalanezhad et al., “Robust voltage control in inverter-interfaced microgrids under plug-and-play functionalities,” IEEE Systems Journal, vol. 14, no. 2, pp. 2813-2824, Dec. 2019. [Baidu Scholar] 

110

C. Mu, Y. Zhang, H. Jia et al., “Energy-storage-based intelligent frequency control of microgrid with stochastic model uncertainties,” IEEE Transactions on Smart Grid, vol. 11, no. 2, pp. 1748-1758, Sept. 2019. [Baidu Scholar] 

111

H. Karimi, M. T. H. Beheshti, A. Ramezani et al., “Intelligent control of islanded AC microgrids based on adaptive neuro-fuzzy inference system,” International Journal of Electrical Power & Energy Systems, vol. 133, p. 107161, Dec. 2021. [Baidu Scholar] 

112

D. Xu, Y. Dai, C. Yang et al., “Adaptive fuzzy sliding mode command-filtered backstepping control for islanded PV microgrid with energy storage system,” Journal of the Franklin Institute, vol. 356, no. 4, pp. 1880-1898, Mar. 2019. [Baidu Scholar] 

113

Y. Yin, J. Liu, J. A. Sanchez et al., “Observer-based adaptive sliding mode control of NPC converters: an RBF neural network approach,” IEEE Transactions on Power Electronics, vol. 34, no. 4, pp. 3831-3841, Jul. 2018. [Baidu Scholar]