**1. Introduction**

Micro-grid comprises the combination of interconnected loads and distributed energy resources (DER), including energy storage devices and several active loads/prosumers which work as a controlled unit to deliver the electric demand for miniature location. It supplies power generation with tremendous reliability as well as an affirmation to varying loads [1–3]. Fossil fuels and nuclear sources are treated as the traditional energy sources, which provide electricity and are not located closer to the load point. As the conventional energy sources are not environmentally friendly and due to the long-distance transmission, there are considerable power losses that can occur. Therefore, nowadays, renewable energy sources have been given more attention by the researchers and industry to generating alternative power [4–12]. Distributed generating (DG) source such as solar, wind, fuel

cell, hydro, tidal, etc. are considered as the main renewable technology, which is highly flexible, expandable and has environmentally friendly behavior. The maximum power point tracking (MPPT) is the important constituent needed to achieve the maximum power point (MPP) as an operating point which enables the utmost power extraction for renewable sources [13,14]. Several MPPT techniques, including Perturb & Observe (P&O), Incremental Conductance (INC), Fuzzy logic control (FLC), Artificial Neural Network (ANN), Particle swarm optimization (PSO), Ant Colony Optimization (ACO), Artificial Bee Colony (ABC), Firefly Algorithm (FA) etc. reviewed in the literature were unable to detect global peak point with partial shade situations [15–27].In this work, Modified Power Ratio Variable Step (MPRVS) based on the P&O technique is proposed without the proportional-integral(PI) controller utilization, which reduces power oscillation near to MPP in comparison to a conventional P&O algorithm and also provides the prevention to battery charging from voltage fluctuation.

To avoid multi reversal generation occurrence in a micro-grid system, in the current research, a Quasi Z-Source inverter is employed [28–33]. The DC-DC converter is a vital interface to achieve a peak power generation from PV modules. In this work, a high-quality tracking behavior is achieved by employing single ended primary inductance converter (SEPIC), which provides high voltage gain with better buck/boost performance compared to other dc-dc switched power converters [34]. In this paper, an additional dc-dc converter (SEPIC converter) is used because it comprises buck/boost capabilities. Moreover, QZsi combines a boost converter and an inverter. The MPRVS based P&O MPPT is controlled through the SEPIC converter which provides MPP achievement and works effectively under varying sun insolation and wind velocity. Moreover, the SEPIC converter works as an impedance adapter between the PV panel and Z-source inverter. Jain et al. [35] have implemented QZsi based grid PV system using a predictive controller in which the active and reactive power have been regulated. However, this work is discussed only for the PV system which utilized the traditional INC MPPT with a classical PI controller as a dc bus regulator. Liu et al. [36] have discussed QZsi based multilevel inverter for grid PV power system, which provides precise MPPT and dc-link voltage regulation at the unity power coefficient. However, during practical justification, voltage/current sensors and bulk resistor models are required, which has a high cost. Nevertheless, this work only explains the performance of QZsi based multilevel inverter for only the PV systems rather than the hybrid system. Amini et al. [37] have discussed the cloud computing applications in micro grid clusters. A real time digital simulator is employed for the physical interpretation of power routing which can be utilized for electrical grid utility with the communication system. However, the application of the proposed scheme with hybrid PV-Wind micro grid systems is missing in this research work. Ali et al. [38] have conferred game theory structure for improvement of smart grid efficiency in which the Femtocell communication system is employed. However, the main disadvantage of this proposed communication system is interference in cross layer. Furthermore, the proposed game theory application with hybrid PV-Wind micro grid system has not been discussed in this research work. Vignesyn et al. [39] have discussed the hybrid micro grid for standalone/Grid mode operation with Z-source inverter. This paper discusses the behavior of micro grid under varying loading conditions, solar insolation and wind speed using simulation environment (MATLAB) only. The real time implementation is missing in this research work. In this research work, to reduce multiple reverse conversions and for improving the efficiency of the micro grid, hybrid PV-Wind with Quasi Z-source inverter has been implemented. Furthermore, SEPIC converter acts as a dc link interface with MPPT functioning. This research work is organized under 3 main sections. Section 1 discusses the micro-grid system with an extensive literature review of MPPT techniques, dc-dc converters with benefits of Z-source inverter. Section 2 presents the complete structure of the hybrid PV-Wind micro-grid system. It explains the PV generator modeling, wind turbine model, MPRVS based MPPT algorithm, design specifications of SEPIC converter, battery model as well as the modes of operations of the Quasi Z-source inverter. Section 3 presents the experimental results which validate the performance of the proposed hybrid PV-wind micro-grid system. The novelty of this research paper is MPRVS based advanced MPPT algorithm have neither

been dis-coursed nor been utilized before for the hybrid PV-wind micro-grid with Quasi Z-source inverter experimentally.

#### **2. Hybrid PV-Wind Micro Grid Structure**

The proposed structure of the PV-Wind micro grid system is shown in Figure 1. The micro grid system contains a PV generator, a Wind Turbine, a battery system and the power electronic converter topologies. To analyze the proposed system, the equivalent circuit with two diode models for the PV generator has been used because of its better power extraction capability when compared with the single diode model. The rotor of the wind turbine is mechanically tied to a generator to produce electrical power. A wind turbine is a complex system, but a reasonably simple representation is possible by modeling the aerodynamic torque or power based on turbine characteristics (non-dimensional curves of the power coefficient). A battery solution is also necessary to balance the stochastic fluctuations of photovoltaic (PV) power and wind power injected to the grid/load. In this section, a short description about how these main components of the proposed micro grid system have been modeled are presented.

**Figure 1.** A block diagram with the structure of Hybrid PV-Wind micro grid system.

### *2.1. PVG Mathematical Model*

Figure 2 illustrates the basic PV cell schematic diagram, which is responsible for the transformation of the solar energy into electric power using photoelectric effect which comprises numerous cells. In this paper, the two-diode model is considered to deliver better accuracy compared to the single diode model.

**Figure 2.** Equivalent circuit model of a PV cell with double diodes and a series and parallel resistance.

The PV cell output current is expressed mathematically as:

$$I\_N = I\_{Photon} - I\_{Diode1} - I\_{Diode2} - \left(\frac{V\_N + I\_N R\_{SE}}{R\_{Parallel}}\right) \tag{1}$$

Also, Photon current is evaluated mathematically as:

$$I\_{Photon} = \left[I\_{Photon\\_STC} + K\_S(T\_C - T\_{STC})\right] \times \frac{G}{G\_{STC}}\tag{2}$$

Diode saturation current can be expressed as:

$$I\_{Diode1} = I\_{Diode2} = \frac{I\_{Short\\_STC} + K\_S(T\_C - T\_{STC})}{\exp\left[\frac{\left(V\_{open\\_STC} + K\_{VL}(T\_C - T\_{STC})\right)}{V\_{Thermal}}\right] - 1} \tag{3}$$

#### *2.2. Wind Turbine Modeling*

A wind turbine is essentially a machine that converts the kinetic energy first into mechanical energy at the turbine shaft, and then into electrical energy. The wind turbine power generation depends mainly on wind velocity in which the rotors are mechanically linked to a generator. A simple model can be achieved by using the power coefficient (CPR) as a function of tip speed ration and the blade pitch angle. CPR (Performance/power coefficient) Vs tip speed (*λT.S*) curve is plotted for different *βP.B* (Pitch blade angle) in Figure 3.

**Figure 3.** CPR (Performance coefficient) Vs tip speed (*λT.S*) curve is plotted for different *βP.B* (Pitch blade angle).

Generated mechanical power output from the wind turbine can be written using Equation (4) which is depending on wind velocity (*VWind*), *RT* (Turbine radius) and *CPR* (Performance coefficient) as:

$$P\_{Mechanical} = \frac{1}{2} \mathbb{C}\_{PR} \pi R\_T^2 \rho\_{a.d} V\_{Wind}^3 \tag{4}$$

Also, the ratio of tip speed (*λT.S*) can be described mathematically which is correlated with an angular velocity of the blade (*ωA.V*), *VWind* and *RT* as:

$$
\lambda\_{T.S} = \frac{\omega\_{A.V} \times R\_T}{V\_{Wind}} \tag{5}
$$

*Energies* **2018**, *11*, 2277

And coefficient of performance is expressed with *λT.S* and *βP.B* (Pitch blade angle) as:

$$\mathbb{C}\_{PR}(\lambda\_{T,S}, \beta\_{P,B}) = 0.72 \left[ \frac{150}{\lambda\_j} - 2 \times 10^{-3} \beta\_{P,B} - 131 \times 10^{-1} \right] e^{\frac{-185 \times 10^{-1}}{\lambda\_j}} \tag{6}$$

where,

$$\frac{1}{\lambda\_{\dot{j}}} = \frac{1}{(\lambda\_{T.S} + 8 \times 10^{-2} \beta\_{P.B})} - \frac{35 \times 10^{-3}}{1 + \beta\_{P.B}^3} \tag{7}$$

$$
\lambda\_{TS} = \frac{\omega\_G \times R\_T}{V\_{\text{Wind}} \times \eta\_{\text{gear}}} \tag{8}
$$

$$
\eta\_{gear} = \frac{\omega\_{GM} \times R\_T}{\lambda\_{T.S} V\_{Wind}} \tag{9}
$$

#### *2.3. Electric Equivalent Circuit of the Battery Model*

A battery is a vital component for a hybrid system which provides the solution under fluctuating action of renewable energy sources. In this work, the electric circuit-based battery model is employed, which provides better dynamics for a state of charge operation mode. It comprises a voltage source (ideal) with a series of internal resistance which evaluates the battery behavior as depicted in Figure 4.

**Figure 4.** Electric equivalent circuit-based battery model.

Final voltage controlled is obtained mathematically as:

$$V = E\_B - \frac{V\_{\rm PO} \times Q\_{\rm Sat}}{Q\_{\rm Sat} - \int I\_{\rm Battery} dt} + A\_{\rm exp} \mathcal{E} \left( B\_{\rm exp} \int I\_{\rm Battery} dt \right) \tag{10}$$
