multilevel converter in a standalone PV system. The system is evaluated under fixed and variable solar

radiation with 15 V as the input voltage 60 V as the required output voltage and 50 kHz as the switching

frequency. Results prove that by using the controller, the system successfully tracks the MPPT point for

constant and variable radiation without oscillation around the maximum power point. In addition, the overshoot

and time response is reduced while the voltage ripples are eliminated. The proposed controller is verified

through practical implementation in Arduino mega board to test the accuracy of results. The practical finding

become essential to achieve the energy transition

and to respond to the continuously increased

consumption of electrical energy based on fossil fuel

sources. However, fossil fuel potential is steadily

decreasing which reinforces the research to find

efficient solutions. The integration of RES to power

different sectors including building and industry is

still limited due to several issues namely, the low

efficiency of renewable energy systems(RESs) and

its intermittency. For these reasons, power

two main categories of converters: non-isolated and

isolated converters. On one hand, the non-isolated

converters are transformer-less converters including

coupled inductor boost converters, interleaved

converters, and integrated boost converters which

capacitor voltage helping to protect the nonlinear

source, [10]. In addition to these advantages, the

SML converter reduces the output voltage transient

overshoot to 46.32 % and minimizes the output

voltage and current ripples to 0.68% and 0.44%

with high efficiency reaching 97% compared to the

classic boost converter, [11], [12]. All of the

control of the SML converter under a closed-loop

mode system. Several controllers are proposed in

the literature to control converters including the

perturb and observe (P&O), fuzzy logic controller,

sliding mode control, and GSS-based MPPT

control, [9], [13], [14], [15]. However, the control of

multilevel converters is still presenting a crucial

issue prohibiting the use of multilevel converters in

several applications. In addition, in the literature,

most papers focus on the classic boost converter as a

improvement and controller robustness. This paper

uses a non-classic boost converter topology as a

conversion device, the symmetrical multilevel

converter, and improves its performance response

through the design of a new controller, a fuzzy

logic-based MPPT controller, to be used in

standalone PV systems. The system performance is

tested under constant and variated radiation using

the Matlab/ Simulink software. In addition, the

paper aims to validate the simulation results via the

The standalone system consists of a solar PV

Where Is represents the saturation current, q is

output. The load is linked differentially to the upper

and lower floating outputs. The diagram of SML is

shown in Figure 3.

non-linearity of the converter, it is difficult to find

advantages: its resilience compared to traditional

nonlinear controllers, its ability to operate with

imprecise inputs, and the management of

nonlinearity; in addition, the FLC do not require an

exact mathematical model to control the converter.

It works based on three steps: The inference engine

with a rule base, the defuzzifier at the output

terminal, and the fuzzifier unit at the input terminal,

[18], [19].

degree of the function to which it belongs. This

mapping is performed according to conditions

given by the rule base. For each linguistic term

that applies to the input variable, a degree of

membership exists.

defined by the user to produce the final signal

according to its comprehension of the system

behavior. The rules used in fuzzy logic

controllers are generally "if-then" statements,

executes these rules and issues an output

variation of the duty cycle, i.e. (∆d) to get

finally a control signal. Indeed, the Mamdani

method is used to find the output of the

inference. each of the inputs admits a boolean

variation as follows:

The variation of the power source (dP/dV)

admits five linguistic variables which are

(Negative Big, Negative Small, Zero, Positive,

and Very Positive) while its derivative has three

The following flowchart, shown in Figure 6,

MPPT controller. The simulation is done using

Matlab/Simulink software as shown in Figure 7.

At the first stage, the solar radiation is fixed to 1000

converter with 15 V as Vpv. The output voltage of

the SML converter reaches 66.46 V as shown in

Figure 8. The system design in closed loop mode

radiation: Pout, and Ppv

Figure 10 illustrates the output and input power

of the system. The first one reaches 745 W as the

average value while the second is 1158 W which

represents 96% of the PV system MPP (PMPPT= 1200

maintained at its maximum power point of 96%.

However, the power loss equals 412.7 W which

and removed using this type of controller. Since the

output voltage is constituted of the input voltage and

capacitor voltages sum. The enhancement of the

capacitor voltage quality by removing the ripples will

surely enhance the output voltage quality and

efficiency. As a result, the elimination of ripples for

the output signal reduces the output signal

oscillations which will be highlighted in the section

is from 0.075 s to 0.1 s, the solar radiation equals 800

Figure 13 illustrates the variation of the output

feature. When passing to the second phase, the solar

radiation is increased consequently the output voltage

generated is boosted to the new MPPT value which is

66.46 V without oscillation or ripples. Similarly, in

the third phase, when the solar radiation is decreased,

the system succeeded in tracking the MPPT point

with the same quality and characteristics. As a result,

the output voltage is at 53 V.

Fig. 12: Solar radiation variation of the PV panel

Fig. 14: The SML current measurement results using Fuzzy logic MPPT technique under solar radiation

Fig. 15: The SML Capacitor voltage measurement results using Fuzzy logic MPPT technique under solar

radiation variation: VC,2,3,4,5,6

Fig. 16: The SML power measurement results using the Fuzzy logic MPPT technique under solar radiation

variation: Pout, and Ppv

The six capacitor voltage measurements in

optimum functioning during the three phases.

Similarly, the capacitor voltage oscillation and

the solar radiation change. Starting from the first

stage, the output power is boosted from 0 W to 588

W without oscillation during the transient regime or

around the maximum power point. In the same

manner, during the second phase where the radiation

and the power of the solar panel are changed, the

output power of the system is increased to reach

741.3 W. Finally, in the third phase, the MPPT is

In this subsection, the testing of the fuzzy logic

adaptive MPPT control in real-time with an Arduino

Mega board and a low-cost co-simulation processor

is implemented. The fuzzy logic adaptive MPPT

control is coded in the Arduino Mega which means

The PIL test gathers the hardware

implementation for the controller and the simulation

for the PV systems and converter. As a result, the

applicability of the systems concerns the validation of

its control using low-cost simulation.

The PIL test can be configured using the Arduino

file Figure 17 (b),

The PIL response during constant and variable

radiation demonstrates compatibility with the

simulation results found in Figure 8 and Figure 13

concerning the output voltage illustrated in Figure 18

and Figure 19 respectively, which proves the