 Original research
 Open Access
 Published:
Robust backstepping global integral terminal sliding mode controller to enhance dynamic stability of hybrid AC/DC microgrids
Protection and Control of Modern Power Systems volumeÂ 8, ArticleÂ number:Â 8 (2023)
Abstract
In this paper, a Backstepping Global Integral Terminal Sliding Mode Controller (BGITSMC) with the view to enhancing the dynamic stability of a hybrid AC/DC microgrid has been presented. The proposed approach controls the switching signals of the inverter, interlinking the DCbus with the ACbus in an AC/DC microgrid for a seamless interface and regulation of the output power of renewable energy sources (Solar Photovoltaic unit, PMSGbased wind farm), and Battery Energy Storage System. The proposed control approach guarantees the dynamic stability of a hybrid AC/DC microgrid by regulating the associated states of the microgrid system to their intended values. The dynamic stability of the microgrid system with the proposed control law has been proved using the Control Lyapunov Function. A simulation analysis was performed on a test hybrid AC/DC microgrid system to demonstrate the performance of the proposed control strategy in terms of maintaining power balance while the systemâ€™s operating point changed. Furthermore, the superiority of the proposed approach has been demonstrated by comparing its performance with the existing Sliding Mode Control (SMC) approach for a hybrid AC/DC microgrid.
1 Introduction
With everincreasing electricity demand, dwindling fossil fuel reserves, and the Greenhouse Gas (GHG) impact on electrical power systems, green Renewable Energy Source (RES) has emerged as the preferred alternative to nonrenewable energy sources or fossil fuels [1, 2]. The Distributed Generation (DG) concept has played a significant role in the transition from conventional fossil fuelbased energy sources to cleaner RES over the last two decades [3]. RES, such as Solar Photovoltaic (PV) and wind energy, are abundant around the world, and they are regarded as the key contributors to improving carbon sustainability [4, 5]. And it is commonly adopted alongside an Energy Storage System (ESS) to form a Microgrid (MG) and serve local electricity demand. The primary benefits of an MG includes to increase dependability, autonomous control, and the flexibility to satisfy local load needs in both islanded (without grid support) and gridconnected modes [6, 7].
However, the intermittencies associated with the RES and the fluctuations in load demands are considered the major issues for MG operators [8, 9]. For example, the generation of wind and solar energy systems, which are the most common RES that features in the MG, depends on the weather conditions (solar irradiation and wind speed). In order to mitigate the intermittencies associated with RES and best utilize their benefits, the Battery ESS (BESS) has become an indispensable component in an MG system [9]. Solar energy and BESS primarily deal with DC electricity, which can be directly used to power DC loads [10, 11]. This will simplify the control structure because the frequency and reactive power regulation are not required in DC operation [12, 13, 26]. Similarly, RES that generates AC power can be utilized to power the AC loads, obviating the need for rectifiers to power DC loads [14]. As a result, hybrid AC/DC MGs with RESs are quickly increasing and capturing the attention of many researchers [15, 16]. An AC/DC MG combines the advantages of both the AC and DC MGs [17, 18], which can handle both the AC and DC loads without the requirement for further ACtoDC and DCtoAC conversions. Furthermore, for a hybrid AC/DC MG, when both AC and DC MG are connected with a Bidirectional Voltage Source Converter (BVSC) reduces the number of conversion steps while simplifying control operations in the MG [13, 19, 26]. However, dealing with intermittencies in RES and fluctuations in loads, and the energy management of diverse components is regarded as a key challenge for the MG operator. Therefore, robust controllers are essential for hybrid AC/DC MGs to provide resiliency against the aforementioned challenges.
Different linear and nonlinear controllers for controlling individual DC and AC MGs have been designed and implemented in the existing literature [20,21,22,23]. The overall aim of these controllers is to maintain the dynamic stability of a MG regardless of whether they are DC or AC. Liu, et al., [23] and Ma et al., [24], identified that a model free ProportionalIntegral (PI) controller can be used to manage the output power of each of the components in hybrid AC/DC MGs where power regulation is achieved by coordinating the actions of individual component controllers. However, the PI controller is not capable to ensure the dynamic stability of hybrid AC/DC MGs under severe system transients. Sowmmiya and Govindharajan, [25] proposed a similar controller for a hybrid AC/DC MG to manage the grid voltage and frequency. In contrast, a distributed control system is proposed in [26] to coordinate the Multiple Parallel Bidirectional Power Converter (MPBPC) in a hybrid AC/DC MG. However, the dynamical stability of DCand ACbus voltages has not been considered in the control approach in [26], though the regulation of these voltages is essential for the desired powersharing in the MG.
Loh et al. [27] proposed a generalized droop control strategy that assures the dynamic stability of the DCbus voltage in a hybrid AC/DC MG while ensuring acceptable powersharing. However, Loh et al. [27] did not consider the powersharing between the AC and DC sides of the hybrid MG. Moreover, it is wellknown that linear controllers suffer from performance deficiencies while seeking to reduce quick transients, such as an abrupt change in generation, a sudden shift in load, a shortcircuit fault, etc. To overcome this drawback, an improved droop control approach has been proposed in [28] to derive the switching signals for the bidirectional VSC, allowing power to be exchanged between the AC and DCsides of a hybrid MG. The fundamental issue with such droop controllers is their sensitivity to voltage fluctuations at the DCor ACbuses of the MG where components of the MG are connected. Furthermore, as droop controllers in a hybrid AC/DC MG are developed using linear characteristics of individual MG components, there is a greater risk of such voltage fluctuations. Also, the unpredictability of RES causes the MG to operate over a wide spectrum. Therefore, the linear droop controllers are incapable of maintaining the dynamic stability of a hybrid AC/DC MG due to their wide operating range.
Besides linear controllers, several nonlinear control techniques for enhancing the transient stability of hybrid AC/DC MGs have been proposed in the current literature [29,30,31]. The nonlinear controllers are expected to provide better performance as compared to linear controllers since these controllers are developed based on the nonlinear dynamical models of a hybrid AC/DC MG, which can describe the system dynamics under a wide range of operating conditions [24]. Adaptive Neural Network (ANNbased controllers have been presented in [32,33,34], to capture nonlinearities associated with different components of a hybrid AC/DC MG. However, the performance of this ANN based controller is dependent on the adequate amount and accuracy of the training data set. This issue associated with the training data set can be solved by the application of modelbased control strategies. A model based nonlinear backstepping controller for a hybrid AC/DC MG has been presented in [1, 35], which considers dynamics associated with each of the components in the MG to ensure acceptable dynamic performance under a wide range of operating conditions. Though the controller in [1, 35] can attain the desired power balance, it is vulnerable to external disturbances (EDs), such as generation or load variations, parameter variations, and faults, which are common during the operation of the MG. To alleviate this limitation, Armghan et al. [29, 31] proposed an adaptive backsteppingbased nonlinear control scheme that assures DCbus voltage regulation. However, the proposed approach is only applicable to the islanded mode operation of the MG.
EDs can be included in the dynamic model as unknown disturbances to capture the modeling errors, parametric uncertainties, and measurement noises associated with a hybrid AC/DC MG. Wang et al., [36] presents a nonlinear disturbance observerbased control technique for a hybrid AC/DC MG, where the observer is used to estimate the unknown disturbance and to assist the controller in adjusting the DC bus voltage to the desired value under these disturbances. However, the control actions are derived using PI scheme, which is expected to be compromised due to the estimation error of the disturbance observer.
As discussed in [37, 38], the Sliding Mode Controller (SMC) is well known for its resilience against parametric uncertainty and EDs. A nonlinear SMC with droop characteristics has been proposed in [39], where the primary goal was to regulate the grid voltage and current considering only the dynamics associated with the BVSC. The presented control approach does not include the dynamics associated with the RES and BESS, even though they have a major influence on the overall dynamic stability of a hybrid AC/DC MG. Considering these facts, Baghaee et al. [40] demonstrated a nonlinear SMC scheme for all the associated components in a hybrid AC/DC MG. Though the SMC approach can properly accomplish DCbus voltage regulation, it exhibits an unexpected chattering issue. To address the chattering issue, Armghan et al. [29, 31] explored a doubleintegral based SMC for a PV and windbased gridconnected hybrid AC/DC MG without considering any ESS. Furthermore, Armghan et al. [29, 31] did not consider neither parametric nor external uncertainties. To overcome this drawback, Hassan et al., [41] designed an adaptive Lyapunov redesign strategy for power converters connected to RES and hybrid ESS considering uncertainties associated with the system. Though it ensures DCbus voltage regulation and energy management, it does not consider the dynamics of the associated DCâ€“AC converter.
Based on the presented literature review, it is evident that a hybrid AC/DC MG with different types of RESbased distributed energy sources, ESS, and associated AC/DC loads are difficult to maintain within the desired operating boundary. With a view to addressing this challenge, this paper proposes a Backstepping Global Integral Terminal Sliding Mode Controller (BGITSMC) for a hybrid AC/DC MG with RES and ESS incorporating EDs associated with MGs, which builds on the approach in [36]. The proposed BGITSMC approach assures the overall dynamic stability of the hybrid AC/DC MG by deriving control inputs in such a way that assures all the associated system states of the MG converge to their intended values. EDs have been incorporated into the dynamical models of various components in hybrid AC/DC MGs, and the proposed control is designed considering the effects of these disturbances. Overall dynamic stability of the MG with the proposed controller has been demonstrated using the control Lyapunov function. The superior performance of the proposed control approach has been demonstrated through rigorous simulation studies and compared with that of the existing SMC (ESMC).
2 Brief overview of a hybrid AC/DC MG
A schematic diagram of a hybrid AC/DC MG is presented in Fig.Â 1. As illustrated in the figure, a hybrid AC/DC MG comprises a DC part and an AC part. The DC part includes all the DC components: DC energy source (solar energy source, PMSGbased wind energy sources), BESS, DC loads, and associated power electronic device [DCâ€“DC Boost Converter (DBC)], and forms a DC MG. On the other hand, the AC part includes the AC loads, which are interfaced with the DC side using a BVSC. Due to the influence of weather intermittencies on the generation of the RES and the load variation, a hybrid AC/DC MG exhibits transients if there is any variation in its operating condition. The control actions associated with the controllable entities in the MG are required to be adjusted to deal with such transients and ensure the dynamic stability of the MG. For example, the DBCs associated with the solar PV units in a MG are required to be regulated to manage output power while assuring the extraction of maximum possible power under varying weather conditions. Similarly, the associated PMSGâ€™s switching control inputs are regulated to extract the maximum output power from the wind farm, while the DCDC Bidirectional Converter (DBBC) associated with the BESS is regulated to control its charging and discharging current. Also, the switching signal driving the BVSC interfacing the DC and AC sides of the MG must also be adjusted according to the intended power exchange between the two sides of the MG to maintain its dynamic stability under varying loads and generation as well as EDs. The accuracy and completeness of the model describing the MG, while deriving its controller. Moreover, the dynamic models of each components of a MG are required to be considered to capture the complete dynamics and to ensure the overall stability of the MG. In the following, first dynamic modeling of the individual components of a hybrid AC/DC MG is developed, followed by the derivation of the proposed controller using the developed dynamic model of the MG.
3 Dynamical model of a hybrid AC/DC MG
This section presents a dynamic model describing each of the components of a hybrid AC/DC MG, which will be used in the following section to develop the proposed controller for the MG. A dynamic model of the BVSC with an output LC filter, which is used to interface the DC and AC sides of the hybrid MG, is presented first. This is followed by a dynamic model of a solar PV unit with DBC, a PMSGbased wind farm with a DBC, the BESS with a DBBC.
3.1 Modeling of the BVSC with an output LC filter
The equivalent circuit of a BVSC with an output LC filter is shown in Fig.Â 2, and the dynamical model of this converter can be developed using circuit theories which are described as follows [22]: with an LC filter.
where \({V}_{dc}\) is the DClink voltage across the capacitor \({C}_{dc}\), \({i}_{o}\) is the current flowing in or out at the DCbus, \({V}_{cd}\) is the directaxis voltage across the filter capacitor (\({C}_{f}\)), \({V}_{cq}\) is the quadratureaxis voltage across the filter capacitor, \(\omega \) is the angular frequency, \({I}_{id}\) is the directaxis current at the input of the LC filter, \({I}_{iq}\) is the quadratureaxis current at the input of the LC filter,\({I}_{od}\) is the directaxis current at the output of the LC filter, \({I}_{oq}\) is the quadratureaxis current at the output of the LC filter, \({L}_{f}\) is the filter inductance, \({M}_{d}\) is the directaxis switching signal of the VSC, and \({M}_{q}\) is the quadratureaxis switching signal of the VSC. By incorporating external EDs into Eq.Â (1), it can be written as:
where \({d}_{1}\), \({d}_{3}\), \({d}_{3}\), \({d}_{4}\), and \({d}_{5}\) are EDs for the BVSC with an output LC filter and the proposed control approach will be used to obtain the switching control inputs, \({S}_{d}\) and \({S}_{q}\).
3.2 Modeling of the solar PV unit with a DBC
The comparable circuit shown in Fig.Â 3 will be used to generate a dynamical model of a solar PV unit with a DBC, with the PV cell assumed to be a single diode model. The detailed dynamical model can be represented by the equations shown below [1]:
where \({v}_{pv}\), \({i}_{Lpv}\), \({C}_{dc}\), \({i}_{opv}\), and \({\mu }_{pV}\) are the solar PV unitâ€™s output voltage, the current flowing through the internal inductance (\({L}_{Lpv}\)), and resistance (\({r}_{Lpv}\)) of the converter, DCbus capacitance, load current, and control signal, respectively. With the EDs in the PV unit with the DBC, the dynamical model in Eqs.Â (3) can be written as follows:
where \({d}_{6}\), \({d}_{7}\) are the EDs for the DBC in the PV system.
3.3 Modeling of the PMSGbased wind farm
The rectifier is directly linked to the stator in a PMSG based wind farm, as illustrated in Fig.Â 4, and the rectifierâ€™s output is connected to the DCbus through a DBC. The DCbus dynamic, on the other hand, is not considered while building the controller for this PMSGbased wind because it has already been addressed by all other components. The stator current dynamics may be used to illustrate the dynamical model of a PMSGbased wind farm, as shown below [20]:
where \({i}_{sd}\) is a directaxis component of the stator current, \({i}_{sq}\) is a quadratureaxis component of the stator current, \({R}_{s}\) is the stator resistance, \({L}_{s}\) is the stator inductance, \({\omega }_{s}\) is the angular frequency of the synchronously rotating frame, \(\psi \) is the stator flux, \({m}_{sd}\) is the directaxis switching control signal for the rectifier, and \({m}_{sq}\) is the quadratureaxis switching control signal for the rectifier. The AC power from the PMSG is converted to DC power before being fed into the DCbus through a rectifier. The rectified current of a diode rectifier can be computed using an average value model, which is expressed as follows:
where \(i_{R}\) is the rectifier current, \(E\) is the per phase rms value of the PMSG, and \(v_{R}\) is the output voltage of the rectifier. As previously stated, the rectifier's output is fed to the DCbus through a DBC. As a result, the DBC dynamic can be written as follows:
where \({i}_{R}\), \({i}_{oR}\), and \({u}_{R}\) represent the current flowing through the inductance (\({L}_{R}\)), and internal resistance (\({r}_{R}\)) of the converter, load current, and control signal, respectively. When the effects of various environmental conditions are factored into Eqs.Â (7), the following result is obtained:
where \({d}_{8}\) and \({d}_{9}\) are the EDs which are used to capture variations or modeling errors in the relevant equations. Using the proposed robust control technique, this dynamical model is utilized to develop the control input.
3.4 Modeling of the BESS with the DBBC
The equivalent circuit of a BESS along with a converter is shown in Fig.Â 5 and the dynamical model can.
be obtained based on this figure which can be written as:
where \({i}_{bat}\), \({v}_{bat}\), \({L}_{bat}\), \({r}_{bat}\), \({i}_{obat}\), and \({\mu }_{bat}\) are used to represent the output current of the BESS, the terminal voltage of the battery, the inductance of the converter, the internal resistance of the inductance, the load current, and the switching control signal, respectively. The dynamical model of the BESS with a DBBC in Eqs.Â (9) can be represented by integrating external disturbances:
where the EDs \({d}_{10}\) and \({d}_{11}\) represent the fluctuations or modeling mistakes in the appropriate equations. The switching control input for the DBBC is designed using the dynamical model described by Eq.Â (10).
In the following section, these dynamic models describing different components of the hybrid MG are used to develop the proposed controller.
4 Proposed controller design
The switching signals for all the power electronic converters associated with different components of the MG are designed using the proposed BGITSMC. Since the design process is the same for each of the MG components, the detailed controller design process is only presented for the BVSC converter with an output LC filter. While, for all other MG components, only the derived control inputs are shown.
4.1 BGITSMC design for the BVSC
The BGITSMC approach is employed on a BVSC with an output LC filter to coordinate powersharing between the AC and DC buses by allowing all state variables linked to the converter to converge to the appropriate steadystate position. Because the suggested approach evaluates the convergences of all connected state variables, the switching control input will be gathered in stages.

Step 1 If the desired value of \({V}_{dc}\) is \({V}_{dc(ref)}\), then the tracking error can be written as follows:
$${e}_{1VSC}={V}_{dc}{V}_{dc(ref)}$$(11)
The dynamic of this error is
To analyze the convergence of \({e}_{1}\), the CLF can be written as:
Using Eq.Â (12), \({\dot{W}}_{1}\) can be written as follows:
It is critical at this stage to choose a stability function (\(\alpha \)) that corresponds to the state variable (\({V}_{cd}\)) in Eq.Â (14) so that \({V}_{dc}={V}_{dc(ref)}\), i.e., \({e}_{1VSC}\) converges to zero. The following options can be used to do this:
where \({k}_{1}\) is a positive design parameter. Using Eq.Â (15), \({\dot{W}}_{1}\) in Eq.Â (14) can be simplified as follows:

Step 2 In this step, the error can be defined as follows:
Using Eq.Â (2), the dynamic of \({e}_{2}\) can be written as:
If \({V}_{cq(ref)}\) is the reference value of \({V}_{cq}\), the corresponding tracking error (\({e}_{3}\)) can be defined as follows:
and its as:
At this time, the second CLF can be selected as follows:
The derivative of \({W}_{2}\) is expressed as follows:
Now the stabilizing functions for twostate variables \({I}_{id}\) and \({I}_{iq}\) need to be selected and these stabilizing functions are as follows:
with \({k}_{2}\) and \({k}_{3}\) as positive control parameters. EquationÂ (22) can be simplified as:
If \({e}_{1}\), \({e}_{2}\), and \({e}_{3}\) converge to zero, \({\dot{W}}_{2}=0\) which indicates the negative semidefiniteness of \({\dot{W}}_{2}\).

Step 3 Since \({\alpha }_{d}\) and \({\alpha }_{q}\) correspond to states \({I}_{id}\) and \({I}_{iq}\), these can be defined as in terms of their error variables as defined below:
These errors' dynamics can be expressed as follows:
At this point, the global integral terminal sliding surface can be defined as follows:
where \(x\) and \(y\) are odd values that must meet the condition \(1<\frac{x}{y}<2\), while \({\tau }_{1}\) and \({\tau }_{2}\) are positive constants. The dynamic of \({S}_{d}\) and \({S}_{q}\), using the values of \({\dot{e}}_{4VSC}\) and \({\dot{e}}_{5VSC}\), can be written as:
The CLF should be chosen as follows to analyze convergences of all errors connected with the BVSC:
whose derivative can be written as:
For the stability of the entire converter, \({\dot{W}}_{3}\) in Eq.Â (30) needs to be negative semidefinite or negativedefinite i.e., \({\dot{W}}_{3}\le 0\) which will be possible if \({M}_{d}\) and \({M}_{q}\) are selected as follows:
where \({k}_{4}\) and \({k}_{5}\) are positive control parameters.
With the switching control inputs in Eq.Â (31), \({\dot{W}}_{3}\) in Eq.Â (30) can be written as follows:
If the following criteria are met, Eq.Â (32) will be negative semidefinite:
\({d}_{1}\le {F}_{1}\), \({d}_{2}\le {F}_{2}\), \({d}_{3}\le {F}_{3}\), \({d}_{4}\le {F}_{4}\), and \({d}_{5}\le {F}_{5}\).with \({F}_{1}\), \({F}_{2}\), \({F}_{3}\), \({F}_{4}\), and \({F}_{5}\) as known bounds. Therefore, the control law as described by Eq.Â (31) can stabilize the BVSC with an output LC filter. The switching control inputs for the remaining converters can be determined in the same way. However, these design techniques have not been repeated in this paper.
5 Proposed controller for the DBC in conjunction with the solar PV unit
The control input for a DBC with a solar PV unit is calculated using the dynamical model provided by Eq.Â (3). TheÂ control input for the DBC in conjunction with the solar PV unit can be obtained using the BGITSMC approach as follows:
where \({e}_{6}\) and \({e}_{7}\) are error variables; and \({M}_{1}=\frac{1}{{L}_{Lpv}}\left[{v}_{pv}{r}_{pv}{i}_{Lpv}{v}_{dc}\right]\) with \({k}_{6}\) and \({k}_{7}\) as positive control parameters while \({F}_{6}\) and \({F}_{7}\) are known bounds to EDs with \(\left{d}_{6}\right\le {F}_{6}\) and \(\left{d}_{7}\right\le {F}_{7}\).
6 Proposed controller for the DBC in conjunction with the PMSG farm
In this subsection, he control input will be derived from the dynamical model of the DBC in conjunction with the PMSGbased wind farm, which is represented by Eq.Â (8). The control law for the PMSGbased wind farm using the BGITSMC approach can be expressed as follows:
where \(N=\frac{{u}_{R}}{{L}_{r}{V}_{dc}}\left[\left(1{u}_{R}\right){i}_{{L}_{R}}+{V}_{dc}\right]+{\dot{M}}_{2}+{\tau }_{5}{e}_{9}+{\tau }_{6}\frac{x}{y}{e}_{9}{(\int {e}_{9}dt)}^{\frac{x}{y}1}\), \({M}_{2}=\frac{1}{{L}_{r}}\left({r}_{R}{i}_{{L}_{R}}+{V}_{dc}\right)\frac{d{i}_{{L}_{R\left(ref\right)}}}{dt}+{k}_{8}{e}_{8}\),
\({e}_{8}\) and \({e}_{9}\) are error variables with \({k}_{8}\) and \({k}_{9}\) as positive control parameters while \({F}_{8}\) and \({F}_{9}\) are known bounds to EDs with \(\left{d}_{8}\right\le {F}_{8}\) and \(\left{d}_{9}\right\le {F}_{9}\).
7 Proposed controller design for the DBBC in conjunction with the BESS
The switching control input for the DBBC in a BESS is obtained using the dynamical model in Eq.Â (10). The switching law for the bidirectional DCDC converter in a BESS can be written using the BGITSMCÂ approach as follows:
where \({e}_{10}\) and \({e}_{11}\) are error variables; and \({M}_{3}=\frac{1}{{L}_{bat}}({v}_{bat}{r}_{bat}{i}_{bat}\frac{d}{dt}{i}_{bat(ref)}\) with \({k}_{10}\) and \({k}_{11}\) as positive control parameters while \({F}_{10}\) and \({F}_{11}\) are known bounds to EDs with \(\left{d}_{10}\right\le {F}_{10}\) and \(\left{d}_{11}\right\le {F}_{11}\).
8 Controller performance evaluation
The dynamic performance of the hybrid AC/DC MG with the proposed controller has been evaluated in this section. A set of simulation studies have been carried out on the MG shown in Fig.Â 1 using the MATLAB/SIMULINK SimpowerSystem software. Under nominal operating conditions, the DC bus voltage is considered to be 640Â V, and the power rating of the solar PV unit is considered to be 12Â kW under standard test conditions (1Â kW/m^{2} and 25Â Â°C) in the test MG. It is noted here that in this work, the conventional Perturb & Observe (P & O) method is used as a Maximum Power Point Tracking (MPPT) algorithm to extract the maximum power from the solar PV unit. The PMSGbased wind farm is rated at 25Â kW. A Lithiumion battery with a capacity of 150 Ah and a nominal voltage of 300Â V has been considered for the MG. The maximum load demand on the DC side of the MG is considered to be 25Â kW, while the maximum load on the AC side is considered to be 30Â kW. The generalized implementation block structure of the proposed control scheme is presented in Fig.Â 6. The developed controller, as seen in this diagram, employs physical attributes like voltage, current, and so on as feedback from the system, while the system model incorporates EDs such as parameter fluctuations, measurement noises, and modeling errors. The impacts of measurement noises are captured in the form of white Gaussian noises, and the bounds on EDs for individual components are derived based on deviations in parameters from their nominal values, combined with modeling errors. These physical attributes, combined with the ED boundaries and nominal parameters of the individual components, are utilized to create the control signal, which is then employed to assure the systemâ€™s dynamic performance. The control signals for all components of the MG are derived from MATLAB/SIMULINK SimPowerSystems, as illustrated in Fig.Â 1. The switching frequencies of the converters during the simulation study are set to 5Â kHz, while the sampling frequency of the simulation study is considered to be 10Â kHz. The system parameters of the proposed hybrid MG system and the gain parameters of the designed controller are mentioned in Tables 1 and 2, respectively.
The DC part of the hybrid AC/DC MG is assumed to run initially in this scenario, with the DCbus voltage being produced by the first activation of the solar PV unit. The PMSGbased wind farm is then put into service, followed by the BESS and the DC load. Once the DC component is completely operational, AC loads are turned on. By evaluating a variety of operating scenarios, the performance of the developed BGITSMC is compared to that of a typical SMC as reported in [39]. It should be mentioned that the existing SMC is designed by considering an integral sliding surface along with the inclusion of a conventional reaching law.
During the simulation study, from time tâ€‰=â€‰0 to 2Â s, the irradiation is considered to be 880 W/m^{2}, which causes a solar PV power output of 9.52Â kW. The PMSGbased wind farm is considered to produce 13.7Â kW of power during this time period, while operating under a wind speed of 9Â m/s. The load demand during this period is considered to be 5.85Â kW DC load and 10.44Â kW AC load, respectively. During the time period tâ€‰=â€‰0â€“2Â s, the overall power generation on the DCside is 23.22Â kW, while the DC load demand is 5.85Â kW. This indicates that there will be a surplus generation of 17.37Â kW of DC power. Depending on the SOC of the battery and the loads in the system, the additional power from the DC side can be either stored or transferred to the AC side. Since the AC load requirement during the time period tâ€‰=â€‰0â€“2Â s is 10.44Â kW, this load demand is met by importing power from the DC side by modulating the bidirectional VSC's switching control signal. The battery stores the energy equivalent to the remaining residual power on the DCside because it still has the potential to store energy. However, the battery charging power is 4.098Â kW, i.e., the power loss is 2.8Â kW. FigureÂ 7 shows the breakdown of the power in different components of the MG during this time period. The DClink voltage is shown in Fig.Â 8, which is maintained at its desired value. As can be seen from Figs. 7 and 8, the performance of the proposed BGITSMC and the ESMC are the same during this time period, as there is no transient event during this period.
In order to demonstrate the performance of the proposed controller at tâ€‰=â€‰2Â s, the following change in condition is considered:

The power generation from the solar PV unit decreases from 9.52 to 7.98Â kW as shown in Fig.Â 7a.
All other conditions, however, remain the same as in the time period tâ€‰=â€‰0â€“2Â s until tâ€‰=â€‰4.5Â s. During this time period, the DCside continues to generate more power than the load demand, while the ACside continues to have a shortfall in power generation. The generated from the DC side, on the other hand, is exported to the AC side to maintain power balance. The battery will now retain less power since the overall power surplus is lower than in the previous stage, as seen in Fig.Â 7e. Transients disrupt the power profiles of different components and the DClink voltage in a hybrid AC/DC microgrid at tâ€‰=â€‰2Â s, as illustrated in Figs. 7 and 8, and the designed BGITSMC handles these transients better (in terms of overshoot and settling time) than the ESMC. As shown in Fig.Â 7d, the AC load changes from 10.44 to 15.4Â kW at tâ€‰=â€‰4.5Â s. All remaining components' power profiles are comparable to those of the prior operating state, i.e., from tâ€‰=â€‰4.5Â s to tâ€‰=â€‰5Â s.
The hybrid AC/DC microgrid experiences another transient event from tâ€‰=â€‰5â€“6Â s due to the following changes:

As shown in Fig.Â 7a, the sun irradiation decreases from 700 to 500 W/m^{2}, causing the output power of the solar PV unit to decrease from 7.98 to 5.42Â kW.
All other conditions remain unchanged. During this time period, the bidirectional VSC will export the same amount of power from the DC side to the AC side as previously explained. However, as total generation is less than total load demand, the energy released by the battery is used to maintain power balance, as illustrated by the BESS power profile in Fig.Â 7e.
The power profiles of the hybrid AC/DC microgrid are disturbed at tâ€‰=â€‰6Â s due to the following event:

As illustrated in Fig.Â 7b, the wind speed increases from 9 to 13Â m/s, raising the output power of the PMSGbased wind farm from 13.7 to 18.5Â kW and

As seen in Fig.Â 7c, the DC load increases from 5.85 to 16.09Â kW.
Other operating conditions remain unchanged as those of the previous time period (i.e., from tâ€‰=â€‰5Â s to tâ€‰=â€‰6Â s) and until tâ€‰=â€‰6.5Â s. The overall generation on the DC side is 23.9Â kW during this time period, whereas the load demand is 16.09Â kW. At the same time, the ACside load requirement is 15.4Â kW. At this moment, to maintain the power balance, the battery discharges its power as shown in Fig.Â 7e. In comparison to all other scenarios, the DClink voltage is drastically disrupted during these transients. However, as compared to the ESMC, the proposed BGITSMC can effectively stabilize this. The AC load is reduced from 15.40 to 10.44Â kW at tâ€‰=â€‰6.5Â s while all reaming components run with power profiles comparable to the preceding phase, and this operation is monitored until tâ€‰=â€‰8Â s.
Finally, at tâ€‰=â€‰8Â s, the solar PV unit power changes from 5.41 to 8.5Â kW, while all reaming components operate with comparable power profiles to the previous period, and this operating condition remains unchanged until tâ€‰=â€‰10Â s. The bidirectional VSC transfers 10.44Â kW of power from the DCside to the ACside in this scenario, and the BESS releases less energy than in the previous time period. Because the intensity of this transient is minor, it has little effect on the power profiles of different components, as seen in Fig.Â 7. However, as seen in Fig.Â 8, the DClink voltage is marginally influenced, but the effect is significantly less severe than under normal conditions. Tables 3, 4, 5, 6, 7 and 8 display the overshoot/undershoot and settling time for the solar PV unit power, PMSG power, DC load power, AC load power, BESS power, and DCbus voltage, respectively under various transient scenarios for both controllers. The designed controller outperforms the existing controller in terms of overshoot and settling time, as shown by the quantitative findings shown in Tables 3, 4, 5, 6, 7 and 8.
The efficacy of the developed BGISMC to maintain the dynamic stability of the hybrid AC/DC MG by balancing power is clearly demonstrated by simulation results under various transient events. This demonstrates the resiliency of the proposed controller against any variation in MG operating conditions.
9 Conclusion
A robust backstepping global integral terminal sliding mode control approach to ensure the dynamic stability of hybrid AC/DC MGs under varying operating conditions has been presented in this paper. The control system derives control inputs for different components of the hybrid MG, while ensuring convergence of all the MG states to their desired values. The rate of change of energy associated with distinct states is observed, and the dynamic stability is theoretically studied using control Lyapunov functions. External disturbances are characterized in terms of parametric uncertainties, modeling errors, and external disturbances that are constrained in such a manner that the proposed controllers maintain dynamic stability in every situation. To support the theoretical conclusions, simulation tests were undertaken, and it was discovered that the developed controllers increase the dynamic stability of hybrid AC/DC MGs by reducing overshoots and settling periods compared to existing approaches. However, certain problems arise during the implementation of the planned controller, which are mostly related to the controller's gain parameter selection. To avoid these difficulties, artificial intelligence can be integrated with the control strategy created.
Availability of data and materials
Not applicable.
Abbreviations
 BGITSMC:

Backstepping global integral terminal sliding mode controller
 PV:

Solar photovoltaic
 BESS:

Battery energy storage system
 CLF:

Control Lyapunov function
 SMC:

Sliding mode control
 GHG:

Greenhouse gas
 RES:

Renewable energy source
 DG:

Distributed generation
 ESS:

Energy storage system
 MG:

Microgrid
 PI:

Proportionalintegral
 MPBPC:

Multiple parallel bidirectional power converter
 ANN:

Adaptive neural network
 EDs:

External disturbances
 BVSC:

Bidirectional voltage source converter
 DBC:

DCDC boost converter
 DBBC:

DCDC bidirectional converter
 P & O:

Perturb & observe
 MPPT:

Maximum power point tracking
References
Roy, T. K., & Mahmud, M. A. (2017). Active power control of threephase gridconnected solar PV systems using a robust nonlinear adaptive backstepping approach. Solar Energy, 153, 64â€“76.
Singh, B., Pathak, G., & Panigrahi, B. K. (2017). Seamless transfer of renewablebased microgrid between utility grid and diesel generator. IEEE Transactions on Power Electronics, 33(10), 8427â€“8437.
Magdy, G., Mohamed, E. A., Shabib, G., Elbaset, A. A., & Mitani, Y. (2018). Microgrid dynamic security considering high penetration of renewable energy. Protection and Control of Modern Power Systems, 3(1), 1â€“11.
Roy, T. K., Mahmud, M. A., Oo, A. M. T., Haque, M. E., Muttaqi, K. M., Mendis, N. (2016) Nonlinear adaptive backstepping controller design for controlling bidirectional power flow of BESSs in DC microgrids. In 2016 IEEE industry applications society annual meeting. IEEE pp. 1â€“8
Papari, B., Edrington, C. S., Bhattacharya, I., & Radman, G. (2017). Effective energy management of hybrid ACâ€“DC microgrids with storage devices. IEEE Transactions on Smart Grid., 10(1), 193â€“203.
Ding, G., Gao, F., Zhang, S., Loh, P. C., & Blaabjerg, F. (2014). Control of hybrid AC/DC microgrid under islanding operational conditions. Journal of Modern Power Systems and Clean Energy., 2(3), 223â€“232.
Firuzi, M. F., Roosta, A., & Gitizadeh, M. (2019). Stability analysis and decentralized control of inverterbased ac microgrid. Protection and Control of Modern Power Systems., 4(1), 1â€“24.
Roy, T. K., Mahmud, M. A., Oo, A. M. T., Haque, M. E., Muttaqi, K. M., & Mendis, N. (2018). Nonlinear adaptive backstepping controller design for islanded DC microgrids. IEEE Transactions on Industry Applications., 54(3), 2857â€“2873.
Liu, S., Zhou, C., Guo, H., Shi, Q., Song, T. E., Schomer, I., et al. (2021). Operational optimization of a buildinglevel integrated energy system considering additional potential benefits of energy storage. Protection and Control of Modern Power Systems., 6(1), 1â€“10.
Ghosh, S. K., Roy, T. K., Pramanik, M. A. H., Sarkar, A. K., Mahmud, M., et al. (2020). An energy management systembased control strategy for DC microgrids with dual energy storage systems. Energies, 13(11), 2992.
Mahmud, M. A., Roy, T. K., Saha, S., Haque, M. E., & Pota, H. R. (2019). Robust nonlinear adaptive feedback linearizing decentralized controller design for islanded DC microgrids. IEEE Transactions on Industry Applications., 55(5), 5343â€“5352.
Chen, D., & Xu, L. (2012). Autonomous DC voltage control of a DC microgrid with multiple slack terminals. IEEE Transactions on Power Systems., 27(4), 1897â€“1905.
Loh, P. C., Li, D., Chai, Y. K., & Blaabjerg, F. (2012). Hybrid ACâ€“DC microgrids with energy storages and progressive energy flow tuning. IEEE Transactions on Power Electronics., 28(4), 1533â€“1543.
Alnejaili, T., Drid, S., Mehdi, D., ChrifiAlaoui, L., Belarbi, R., & Hamdouni, A. (2015). Dynamic control and advanced load management of a standalone hybrid renewable power system for remote housing. Energy Conversion and Management., 105, 377â€“392.
SimpsonPorco, J. W., DÂ¨orfler, F., & Bullo, F. (2016). Voltage stabilization in microgrids via quadratic droop control. IEEE Transactions on Automatic Control., 62(3), 1239â€“1253.
Gupta, A., Doolla, S., & Chatterjee, K. (2017). Hybrid ACâ€“DC microgrid: Systematic evaluation of control strategies. IEEE Transactions on Smart Grid., 9(4), 3830â€“3843.
Sun, Q., Zhou, J., Guerrero, J. M., & Zhang, H. (2014). Hybrid threephase/singlephase microgrid architecture with power management capabilities. IEEE Transactions on Power Electronics., 30(10), 5964â€“5977.
Nejabatkhah, F., & Li, Y. W. (2014). Overview of power management strategies of hybrid AC/DC microgrid. IEEE Transactions on Power Electronics., 30(12), 7072â€“7089.
Li, P., & Zheng, M. (2019). Multiobjective optimal operation of hybrid AC/DC microgrid considering sourcenetworkload coordination. Journal of Modern Power Systems and Clean Energy., 7(5), 1229â€“1240.
Chinchilla, M., Arnaltes, S., & Burgos, J. C. (2006). Control of permanentmagnet generators applied to variablespeed windenergy systems connected to the grid. IEEE Transactions on Energy Conversion., 21(1), 130â€“135.
Ackermann, T. (2005). Wind power in power systems (Vol. 200). Wiley.
Nguyen, H. T., Kim, E. K., Kim, I. P., Choi, H. H., & Jung, J. W. (2017). Model predictive control with modulated optimal vector for a threephase inverter with an LC filter. IEEE Transactions on Power Electronics., 33(3), 2690â€“2703.
Liu, X., Wang, P., & Loh, P. C. (2011). A hybrid AC/DC microgrid and its coordination control. IEEE Transactions on Smart Grid., 2(2), 278â€“286.
Ma, T., Cintuglu, M. H., & Mohammed, O. A. (2016). Control of a hybrid AC/DC microgrid involving energy storage and pulsed loads. IEEE Transactions on Industry Applications., 53(1), 567â€“575.
Sowmmiya, U., & Govindarajan, U. (2017). Control and power transfer operation of WRIGbased WECS in a hybrid AC/DC microgrid. IET Renewable Power Generation., 12(3), 359â€“373.
Xia, Y., Peng, Y., Yang, P., Yu, M., & Wei, W. (2016). Distributed coordination control for multiple bidirectional power converters in a hybrid AC/DC microgrid. IEEE Transactions on Power Electronics., 32(6), 4949â€“4959.
Loh, P. C., Li, D., Chai, Y. K., & Blaabjerg, F. (2012). Autonomous operation of hybrid microgrid with AC and DC subgrids. IEEE Transactions on Power Electronics., 28(5), 2214â€“2223.
Wang, J., Jin, C., & Wang, P. (2017). A uniform control strategy for the interlinking converter in hierarchical controlled hybrid AC/DC microgrids. IEEE Transactions on Industrial Electronics., 65(8), 6188â€“6197.
Armghan, H., Yang, M., Wang, M., Ali, N., & Armghan, A. (2020). Nonlinear integral backstepping based control of a DC microgrid with renewable generation and energy storage systems. International Journal of Electrical Power & Energy Systems., 117, 105613.
Li, P., Guo, T., Zhou, F., Yang, J., & Liu, Y. (2020). Nonlinear coordinated control of parallel bidirectional power converters in an AC/DC hybrid microgrid. International Journal of Electrical Power & Energy Systems., 122, 106208.
Armghan, H., Yang, M., Armghan, A., Ali, N., Wang, M., & Ahmad, I. (2020). Design of integral terminal sliding mode controller for the hybrid AC/DC microgrids involving renewables and energy storage systems. International Journal of Electrical Power & Energy Systems., 119, 105857.
Akpolat, A. N., Habibi, M. R., Baghaee, H. R., Dursun, E., Kuzucuoglu, A. E. E., Yang, Y. (2021) Dynamic stabilization of DC microgrids using ANNbased model predictive control. IEEE Transactions on Energy Conversion.
Chettibi, N., Mellit, A., Sulligoi, G., & Pavan, A. M. (2016). Adaptive neural networkbased control of a hybrid AC/DC microgrid. IEEE Transactions on Smart Grid., 9(3), 1667â€“1679.
Kang, K. M., Choi, B. Y., Lee, H., An, C. G., Kim, T. G., Lee, Y. S., et al. (2021). Energy management method of hybrid AC/DC microgrid using artificial neural network. Electronics, 10(16), 1939.
Roy, T. K., & Mahmud, M. A. (2017). Dynamic stability analysis of hybrid islanded DC microgrids using a nonlinear backstepping approach. IEEE Systems Journal., 12(4), 3120â€“3130.
Wang, C., Li, X., Guo, L., & Li, Y. W. (2014). A nonlineardisturbanceobserverbased DCbus voltage control for a hybrid AC/DC microgrid. IEEE Transactions on Power Electronics., 29(11), 6162â€“6177.
Roy, T. K., Mahmud, M. A. (2021). Fault current compensations in resonant grounded distribution systems to mitigate powerline bushfires using a nonsingular terminal sliding model controller. IET Generation, Transmission & Distribution.
Roy, T., Mahmud, M. A., Ghosh, S., Pramanik, M., Kumar, R., Oo, A. M. (2021) Design of an adaptive sliding mode controller for rapid earth fault current limiters in resonant grounded distribution networks to mitigate powerline bushfires. In 2021 IEEE Texas Power and Energy Conference (TPEC). IEEE. p. 1â€“6
Alam, F., Ashfaq, M., Zaidi, S. S., Memon, A. Y. (2016) Robust droop control design for a hybrid AC/DC microgrid. In 2016 UKACC 11th international conference on control (CONTROL). IEEE. pp. 1â€“6.
Baghaee, H. R., Mirsalim, M., Gharehpetian, G. B., Talebi, H. A. (2017) A decentralized power management and sliding mode control strategy for hybrid AC/DC microgrids including renewable energy resources. IEEE Transactions on Industrial Informatics.
Hassan, M., Paracha, Z. J., Armghan, H., Ali, N., Said, H. A., Farooq, U., et al. (2020). Lyapunov based adaptive controller for power converters used in hybrid energy storage systems. Sustainable Energy Technologies and Assessments., 42, 100853.
Acknowledgements
Not applicable.
Funding
Not applicable.
Author information
Authors and Affiliations
Contributions
Conceptualization, TKR and SKG; methodology, TKR and SS; software, TKR and SKG; validation, TKR and SKG; formal analysis, TKR; investigation, TKR; writingâ€”original draft preparation, TKR and SKG; writingâ€”review and editing, TKR, SKG and SS; supervision, TKR and SS; and project administration, TKR and SS. All authors read and approved the final version of the manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Roy, T.K., Ghosh, S.K. & Saha, S. Robust backstepping global integral terminal sliding mode controller to enhance dynamic stability of hybrid AC/DC microgrids. Prot Control Mod Power Syst 8, 8 (2023). https://doi.org/10.1186/s41601023002812
Received:
Accepted:
Published:
DOI: https://doi.org/10.1186/s41601023002812
Keywords
 Dynamic stability
 Hybrid AC/DC microgrids
 Power balance
 Robust backstepping controller
 Global integral terminal sliding mode controller
 Switching control signals