Introduction

In recent decades, the usage of fossil fuels has drastically augmented owing to the mandate for electricity in human day-to-day life1,2. The continued consumption of fossil fuels has led to their depletion, and their combustion produces harmful by-products. Finding the best clean and green energy sources that can keep up with the needs of today's fully electrified society has been a major focus of scientific inquiry for the past few decades3,4. Photovoltaic (PV) and wind are two of the utmost promising clean energy in use today because of their low or no environmental impact and high potential for widespread adoption. The grid installed capacity of the renewable energy sources is 133,886.18 MW, in which PV contribution is 54% and wind contribution is 33% as of the 31st December 20235.

The unpredictable and at times unpredictable input weather conditions of clean energy sources make for erratic power output and an inability to keep up with load demands6. This raises doubts about harnessing the power of renewable. The solution to the aforesaid issues, the incorporation of multiple Regenerative sources with storage devices is one of the most promising solutions. Integration methodologies and maximum power tracking power converters are needed to maximize hybrid system power. The literature contains many power converters for hybridizing energy sources, including AC shunt, DC shunt, and hybrid mixed systems.

In7, the authors developed a high-gain converter for multiple RES. By changing the linked inductor winding turns ratio, the system achieves high voltage transfer gain. The experimental findings are compared with the theoretical values and the implanted hybrid system provides an overall efficacy of 94%.

A novel DC-DC converter of three input ports has been formulated and presented in8, the configuration provides input terminals for solar power and Fuel Cell systems and a bi-directional port for connecting the battery system. Four peculiar duty ratio signals are generated to govern the power switches in the developed configuration.

A new zero voltage switching converter was presented in9, a multi-input converter for boosting the peculiar low voltage to desired and stable voltages. A complementary circuit consisting of negligible inductance is engaged to turn on the converter in two modes.

A flexible matrix converter was invented by10 to add renewable energy sources to the grid. The nine-switch matrix converter works similarly, although it has additional power inputs. The lower section unites three hybrid energy systems, while the above section communicates with the wind power generator. Simple Ex-OR gate logic modulates and controls the system, which features a top half that operates as a 3-Ø rectifier for the wind and a bottom half that provides three sources with DC-DC conversion.

The authors11 developed an integrated Cuk-SEPIC converter by sharing the Cuk inductor L2 with the SEPIC and by rearranging the converter diodes which reduces the converter components for the hybrid system and eliminates the need for the input filter.

The additional major issue is the requirement of individual and dedicated maximum power tracking algorithms, which causes complexity in the system implementation. In literature, many authors have proposed universal MPPT controllers12,13,14,15,16, which are worm to elicit the maximal power from RES, but the universal MPPT techniques have limitations of requiring a dedicated controller for each source, which in turn increases the implementation complexity.

MPPT controllers vary in power extraction, tracking speed, and other features. Authors17,18,19,20,21 suggested maximum power point (MPP) extraction from RES with consistent tracking speed and MPPT controller characteristics to solve system implementation issues. Many MPPT algorithms remove MPP from PV and wind sources22,23,24,25,26.

A hybrid Luo (HL) converter with one MPPT controller is shown in this study. The suggested converter splits charging and DC link capacitors across converters with negative output to produce a multi-input system. The solar-wind energy system may now harvest maximum power points with a unified MPPT controller. A hybrid converter MPPT architecture controls power from both sources better.

In this article, "Design of hybrid system" section presents the design of a proposed hybrid system. In "Analysis of hybrid Luo converter" section, the analysis of the hybrid Luo converter. The Unified MPPT control techniques are discussed in "Unified MPPT control techniques" section. The analysis and discussion of the proposed system are presented in "Analysis and discussion on the proposed system" section. Finally, "Conclusion" section presents the conclusion.

Design of hybrid system

Figure 1 depicts the system with a unified MPPT controller. This system has a Luo DC-DC converter and a controller for MPP extraction from a hybrid RES system.

Figure 1
figure 1

Proposed hybrid system with HL converter and unified MPPT topology.

Full size image

PV system

PV arrays have series and parallel modules. Figure 2 shows the PV cell circuit and symbol.

Figure 2
figure 2

(a) PV cell, (b) symbolic PV cell representation.

Full size image

The PV system's output is determined by the PV module's specification parameters. In the developed system, the BP solar SX3190 PV module is designated to model a 560W system. Table 1 lists the specific parameters and the system's production is dependent on temperature and solar radiation levels.

Table 1 Details of 560 PV system.
Full size table

Wind system

A converter is placed across a wind turbine's generator and the electrical grid or load to optimize wind power. The received power from the system is expressed by Eq. (1)18.

$$P_{m} = \frac{1}{2}\rho AV_{v}^{3}$$
(1)

The generated mechanical power is dependent on the wind turbine friction coefficient, tip speed ratio, air density, and wind velocity, according to Eq. (1). The Aeolos-H 500 W model's default settings are utilized to construct a 500 W wind system for use in a hybrid setup. Details are provided in Table 2.

Table 2 Details of a wind system.
Full size table

Analysis of hybrid Luo converter

A hybrid Luo converter topology is derived in this section for the amalgamation of renewable PV and wind sources with reduced power converter switches and converter components count.

Design of converter

The hybrid Luo (HL) converter in Fig. 3 is based on the super lift Luo converter27.

Figure 3
figure 3

HL converter topology.

Full size image

Negative-output super lift two solar and wind input ports are created by merging super lift Luo converters. In Fig. 3, the charging and capacitors of C1 and C2 are divided to provide the mutually merged configuration. depending on renewable energy sources, the suggested converter should manage four circumstances, which may be attained by delivering the pulse to the control switches of separate dual input converters and interpreted in the following sections.

Operating conditions of HL converter

Condition-1: S1—closed, S2—closed

S1 and S2 are closed in condition 1. Figure 4a shows the sharing capacitor C1 charging with VPV and VW voltage. Equation (2) calculates the inductor currents iL1 and iL2.

$$i_{L1} = I_{w} + \frac{{V_{w} + V_{c1} }}{{L{}_{1}}}t\;\;{\text{and}}\;\;i_{{{\text{L}}2}} = {\text{I}}_{{{\text{pv}}}} + \frac{{{\text{V}}_{{{\text{pv}}}} + {\text{V}}_{{{\text{c}}1}} }}{{{\text{L}}{}_{2}}}{\text{t}}$$
(2)
Figure 4
figure 4

(a–d) Operating conditions of proposed HL converter.

Full size image

Condition-2: S1—closed, S2—open

Condition-2 has S1 closed and S2 open. In Fig. 4b, VW charges C1's sharing capacitor. Equation (3) calculates the inductor currents iL1 and iL2.

$$i_{L1} = I_{w} + \frac{{V_{w} + V_{c1} }}{{L{}_{1}}}t\;\;{\text{and}}\;\;i_{L2} = I_{dc} - \frac{{V_{c1} }}{{L{}_{2}}}t$$
(3)

Condition-3: S1—open, S2—closed

Condition-3 has S1 open and S2 closed as in Fig. 4c, the sharing capacitor C1 is charging through VPV. Equation (4) calculates the inductor currents iL1 and iL2.

$$i_{L1} = I_{dc} - \frac{{V_{c1} }}{{L{}_{1}}}t\;\;{\text{and}}\;\;i_{L2} = I_{pv} + \frac{{V_{pv} + V_{c1} }}{{L{}_{2}}}t$$
(4)

Condition-4: S1—open, S2—open

The switches S1 and S2, are in the open position in condition 4. In this state, the DC link capacitor C2 is being discharged by the distribution capacitor and inductors L1 and L2, which results in the boosted voltage at the load side as shown in Fig. 4d. Equation (5) formulates the currents iL1 and iL2 over L1 and L2.

$$i_{L1} = I_{dc} - \frac{{V_{c1} }}{{L{}_{1}}}t\;\;{\text{and}}\;\;i_{L2} = I_{dc} - \frac{{V_{c1} }}{{L{}_{2}}}t$$
(5)

Table 3 is a list of the operating condition summaries for the proposed HL converter with switching positions.

Table 3 Operating conditions of HL converter.
Full size table

To provide hybrid multi-input power converters with fewer power converter components and to lessen the switching stress of the converter, the HL converter configuration is built by reordering and sharing the converter components between the converters. The proposed converter switching waveforms are shown in Fig. 5.

Figure 5
figure 5

Proposed HL converter switching waveforms.

Full size image

Unified MPPT control techniques

The work aims to extract MPP from dynamically varying RES via maximum power tracking. P&O, Hill climbing, artificial neural networks, fuzzy logic controllers and bio-inspired algorithms are available for PV and wind. Due to its ease of use and predefined step size, the P&O approach is the most popular. Fuzzy and ANN-based MPPT can manage non-linearity.

Unified P&O MPPT controller

A unified algorithm from the classic P&O controller generates a duty cycle for the 560 W HL converter-fed hybrid PV and 500 W wind systems.

Equation (6) calculates PV output power.

$${\text{P}}_{{{\text{PV}}}} = {\text{ V}}_{{{\text{PV}}}} *{\text{ I}}_{{{\text{PV}}}}$$
(6)

Equation (7) shows the DC link voltage-PV output voltage correlation.

$$V_{PV} = \left( {\frac{1}{1 - D}} \right)V_{dc}$$
(7)

Equation (8) calculates wind output power as

$${\text{P}}_{{\text{W}}} = {\text{ V}}_{{\text{W}}} *{\text{ I}}_{{\text{W}}}$$
(8)

The voltage at the DC link and solar power generation voltages are related in Eq. (9).

$$V_{W} = \left( {\frac{1}{1 - D}} \right)V_{dc}$$
(9)

The suggested unified algorithm extracts maximum power from renewable sources in steps. Based on irradiance and velocity, the unified P&O MPPT approach delivers a hybrid system duty cycle. P&O-based control strategies require many assessments, slowing convergence is shown in Fig. 6. Comparing unified MPPT controllers to individual MPPT controllers, the latter provides a more straightforward and economical solution for renewable energy systems. Through full utilization of renewable energy sources, they minimize expenses, simplify system architecture, and enhance overall performance. Unified controllers are a good option for many installations due to their improved coordination, ease of monitoring and maintenance, and space savings; nevertheless, the choice ultimately depends on the particular system requirements and design considerations.

Figure 6
figure 6

Modified P&O MPPT.

Full size image

RBFN MPPT controller

The current section develops a unified ANN-based controller to monitor wind and PV maximum power points. Figure 7 exhibits an RBFN-based network featuring four input variables (VPV, IPV, VW, and IW), 529 concealed neurons in the hidden layer, and two regulatory switch duty cycles. In three steps, the same RBFN-based MPPT controller generates a duty cycle for concurrent renewable energy sources17,18. The nodal equations for RBFN-based single MPPT controllers are as follows:

Figure 7
figure 7

RBFN Unified MPPT controller.

Full size image

Input layer

Input layer nodes immediately send VPV, IPV, VW, and IW data to the next neuron layer. Equations (10) and (11) inure the input layer and total input.

$$x_{i}^{\left( I \right)} \left( n \right) = net_{i}^{\left( I \right)}$$
(10)
$$y_{i}^{\left( I \right)} \left( n \right) = f_{i}^{\left( I \right)} \left( {net_{i}^{I} \left( n \right)} \right) = net_{i}^{\left( I \right)} \left( n \right),^{{}} \quad i = 1,2,3,4$$
(11)

Hidden layer

The membership function for each hidden layer is Gaussian. Equation (12) accounts for all inputs, while Eq. (13) accounts for all outputs.

$$net_{j}^{\left( H \right)} \left( n \right) = - \left( {X - M_{j} } \right)^{T} \mathop \sum \limits_{j} \left( {X - M_{j} } \right)$$
(12)
$$\begin{aligned} y_{j}^{\left( H \right)} \left( n \right) & = f_{j}^{\left( H \right)} \left( {net_{j}^{\left( H \right)} \left( n \right)} \right) \hfill \\ \mathop {}\nolimits_{{}} \mathop {}\nolimits_{{}} \mathop {}\nolimits_{{}}^{{}} & = \exp \left( {net_{j}^{\left( H \right)} \left( n \right)} \right),\mathop {}\nolimits_{{}} j = 1,2,3,4 \hfill \\ \end{aligned}$$
(13)

Output layer

The controller's total output is made up of two neurons in output. These neurons use the linear activation function to create two diverse pulses, Dpv and Dw. In Eqs. (14) and (15), we find the output layer, the total input, and the output.

$$net_{k}^{\left( O \right)} \left( n \right) = \sum\limits_{j} {w_{j} y_{j}^{\left( H \right)} \left( n \right)}$$
(14)
$$y_{k}^{\left( O \right)} \left( n \right) = f_{k}^{\left( O \right)} \left( {net_{k}^{\left( O \right)} \left( n \right)} \right) = net_{k}^{\left( O \right)} \left( n \right)$$
(15)

where, \(x_{i}^{\left( I \right)}\) = input layer;\(net_{i}^{\left( I \right)} \left( n \right)\) = input layer summated value; \(net_{i}^{\left( H \right)} \left( n \right)\) = hidden layer summated value; \(net_{i}^{\left( O \right)} \left( n \right)\) = output layer summated value; Wj = hidden and output layer connecting weight; Mj = output layer’s Mean deviation; \(\sum\limits_{j} {}\) = standard deviation of the output layer.

Currents and voltages of PV and Wind regulate the system outputs, Solar photovoltaic (PV) irradiance and wind speed define each system's duty cycle. The Radial Basis Function Neural Network (RBFN) MPPT controller offers potential advantages over the Perturb and Observe (P&O) MPPT controller, including better adaptability to changing conditions, faster response, reduced oscillations, improved performance under partial shading, enhanced robustness, and the potential for online adaptation. The choice between them depends on factors like implementation details, cost, and real-time processing requirements.

Analysis and discussion on the proposed system

MATLAB/Simulink verifies HL converter topology performance. From the literature, PV and wind energies are contradictory, so to test the converter's feasibility in both exclusive and simultaneous modes, renewable sources are selected for 0–0.3 s, with the wind at 12 m/s and PV at 600 W/m2. As illustrated in Fig. 8, a wind velocity of 10 m/s with 800 W/m2 PV radiation and a velocity of the wind of 8 m/s with 1000 W/m2 are used for the 0.3–0.6 s and 0.6–0.9 s intervals, respectively.

Figure 8
figure 8

(a) Considered solar irradiation, (b) considered wind speed.

Full size image

The proposed hybrid system with a PV power rating of 560 W and wind power rating of 500 W with the developed converter topology detailed particulars are presented in Table 4.

Table 4 Detailed particulars of developed hybrid system.
Full size table

The hybrid system with the designed HL converter and the planned unified MPPT controllers is tested in standalone and grid-connected settings with three load zones based on renewable sources.

Standalone mode

Figure 9 illustrates the functionality of the unified P&O MPPT architecture to demonstrate the voltage, current, and power measured at the DC link capacitor using the unified technique applied to the system. Due to the non-linear features of PV irradiance and Wind speed, the P&O approach is unable to collect the maximum amount of electricity from the sun and wind.

Figure 9
figure 9

Voltage, current, and power at DC link capacitor with single P&O MPPT.

Full size image

Figure 10 demonstrates the voltage, current, and power measured at the DC link capacitor with a unified RBFN-based MPPT topology. RBFN-MPPT measures maximum power and stabilizes during irradiance and wind speed parameter changes. RBFN-based MPPT achieves constant DC link voltage due to its faster convergence speed. DC microgrid efficiency relies on stable DC link voltage.

Figure 10
figure 10

Voltage, current and power at DC link capacitor with unified RBFN MPPT.

Full size image

Table 5 summarizes the hybrid system with updated MPPT control approaches and renewable sources. P&O MPPT generates 754.2 W and RBFN-based MPPT generates 756.7 W for 600 W/m2 and 12 m/s input data from 0 to 0.3 s. For 0.3 to 0.6 s, P&O MPPT develops 778 W and RBFN-based MPPT develops 781.2 W. For 0.6 to 0.9 s, P&O develops 801.2 W and RBFN develops 804.6 W. The table shows that the unified RBFN controller outperforms the unified P&O controller.

Table 5 The summary hybrid system.
Full size table

Experimental results

To validate the contemplated converter, A laboratory-tested prototype model is developed for the scale-down model and it is depicted in Fig. 11.

Figure 11
figure 11

Experimental prototype.

Full size image

MSOs evaluate the outcomes of experiments. Sensors for voltage (LV 20P) and current (LA 25N) from the dSPACE 1104 model allow for feedback control signals from a renewable source.

The PV output power of the converter arrangement is depicted in Fig. 12. The non-linear performance of the system is put to the test by the programmable DC Source's sudden shift in irradiance and wind speed. The most power may be produced from PV irradiation using RBFN-based MPPT.

Figure 12
figure 12

PV output power.

Full size image

Power generated by the wind with this converter setup is depicted in Fig. 13. The recommended layout is supported by the observed step change in the wind. The speed is 12-10-8 m/s. Energy output depends on wind speed, with 453.5 W, 377.7 W, and 299.3 W being generated at various gusts. As a result, MPPT based on RBFN efficiently measures wind velocity.

Figure 13
figure 13

Wind output power.

Full size image

Figure 14 illustrates the anticipated system DC link voltage. The hybrid system balances a constant (228 V) DC link voltage to sustain the DC microgrid. The constant DC connection voltage is the foundation of the RBFN unified MPPT technique.

Figure 14
figure 14

DC link output voltage.

Full size image

Figure 15 displays the hybrid system's overall output power. The inputs of (503.4 w) and (299.2 w) of PV and Wind, respectively, yield the maximum power of (802.6 W).

Figure 15
figure 15

Output power.

Full size image

Grid mode

A MATLAB model of a 560W PV and a 500W wind, along with a suggested HL converter and a unified MPPT controller, are developed in MATLAB with an AC grid that has a rating of 230 V and 50HZ. This is done to test the hybrid system when it is linked to the grid. The following is an analysis of the load that is coupled to the proposed grid-connected system: for the first period, 0 to 0.3 s, the load is considered to be 500W. Similarly, the power output for the second period, which lasted from 0.3 to 0.6 s, was 750 watts, while the power output for the third period, which lasted from 0.6 to 0.9 s, was 1000 watts.

The 3-phase voltage source inverter receives the common DC link voltage based on input source information and delivers sinusoidal pulse width modulation pulses to control it. Figures 16, 17, 18 show inverter, grid, and load voltages and currents.

Figure 16
figure 16

Inverter side voltage and current profile.

Full size image
Figure 17
figure 17

Grid voltage and current profile.

Full size image
Figure 18
figure 18

Load voltage and current profile.

Full size image

In the proposed hybrid system, the load requirement is met by allocation with the grid based on the inverter output power, as shown in Fig. 19 for three different load scenarios. From zero to thirty seconds, the net inverter output is 754.1 W, the load mandate is 500 W, and the surplus of 254.1 W is fed back into the grid. Similarly, between 0.3 and 0.6 s, the load mandate is 750 W while the generated power is 780.4 W; the surplus 30.4 W power is fed to the grid; and between 0.6 and 0.9 s, the load mandate is 1000 W while the generated power is 802.2 W; the remaining 197.8 W power is drawn from the grid to meet the demand on the load side. The peak active power load demand from both the grid and the hybrid system is shown in Table 6.

Figure 19
figure 19

Active load, inverter and grid power profile.

Full size image
Table 6 Summary of load demand with inverter output and grid power.
Full size table

Conclusion

The unified P&O and unified RBFN MPPT controllers are suggested in this work in conjunction with a hybrid Luo converter to build a hybrid RES system. The literature on hybrid energy sources that are sustainable covers a wide range of multi-input DC-DC converters and MPPT methods. A system that is hybrid has been proposed and researched by taking into account both the wind system (500 W) and the PV system (560 W). Taking into account data from an assortment of renewable sources throughout three separate load scenarios in three geographically varied areas, this study examines the built hybrid system's performance both off-grid and in grid-connected modes. The designed system's output is validated by simulation and hardware prototype model findings. In a stand-alone condition, the hybrid system generates an average of 756.7 W using a unified RBFN MPPT controller in the first region with a wind of 12 m/s and PV of 600 W/m2, 781.2 W in the second region with a wind of 10 m/s and 800 W/m2, and 804.6 W in the third region with an 8 m/s wind and 1000 W/m2. Correlations between the obtained findings and the P&O-MPPT controller confirm that the anticipated RBFN-MPPT controller offers the most promising option. This study presents the active powers profile for three distinct loads to verify the proposed controller's operation in a grid-tied setting. The complexity of implementing a hybrid system was reduced thanks to a new MPPT controller based on a modified RBFN.