*3.1. Issues Related to Conventional P&O and INC Algorithms*

The P&O and INC techniques utilize the (*P*-*V*) characteristics of the PV panel in the tracking procedure, fulfilling the requirement *dP/dV* = 0. Essentially, it is difficult to determine the zero point on the *P-V* curve's slope as in P&O and the truncation error in digital processing as in the INC. Therefore, both approaches turned out to be entirely inaccurate in tracing the MPP owing to the resulting steady-state oscillations in the vicinity of the MPP and higher response time. The perception of the P&O algorithm is centered on modulating the duty cycle. If the peak of the *P-V* characteristic is detected, there will be no more additional perturbations to the duty cycle. This process, however, causes the system operation to be adjacent to the MPP, but not at the point itself. Therefore, contentious duty cycle modifications must be employed to maintain the MPP position resulting in oscillations that are proportional to the step size. The larger step size is translated to higher oscillations while slower tracking is the result of the small step size.

In contrast, the concept of the INC algorithm depends on determining the slope of the power curve by employing the incremental conductance of the PV panel. When the incremental conductance has the same value as its instantaneous one, the maximum power can be successfully traced. Unlike the P&O algorithm, the INC process requires a powerful microcontroller which increases the system cost [5]. As a result of these issues raised for both P&O and INC algorithms, they are not the best choice when sudden changes occur in solar irradiance and/or temperature.

#### *3.2. Modified MPPT Algorithms*

Based on the discussion mentioned above, in this subsection, we present modified algorithms to address the issues related to the traditional P&O and INC techniques. The modification is based on the variable-step size as an alternative to using a fixed-step size. Additionally, a comptonization between fast response and steady-state oscillations will be met. The approach of variable step size adjustment was previously outlined in [37,38]. It used a scaling factor reliant on the variation in power (Δ*P*) and voltage (Δ*V*). Nevertheless, this modification may not display good tracing capabilities in abrupt irradiance variation [5]. So, to swamp this problem, we follow the modified algorithms presented in [5] using Matlab-Simulink, depending on an enhanced variable step size that relies solely on the power change (Δ*P*) with a scaling factor accustomed to the settlement of the response time and decreasing the steady-state oscillations in such a way as to reduce the oscillations of the output PV power as given in Figures 3 and 4, which illustrate the flow chart of M-VSS-P&O and M-VSS-INC techniques, respectively. The power (*P*) is calculated from *V* × *I*, while the change in power is calculated as the difference between the new and old values (i.e., Δ*P=Pnew* − *Pold*).

**Figure 3.** Flowchart of modified P&O algorithm.

**Figure 4.** Flowchart of Modified INC algorithm.

It is well known, for the conventional P&O algorithm, that when irradiance increases, a drift problem occurs, which causes a delay in tracing the MPP. This is indicated in Figure 5, which displays the *P-V* curves for two cases of irradiance. If the operating point is at A, and there is an abrupt increase in insolation and the point will settle to point B. At point B, Δ*P* > 0 and Δ*V* > 0, so the algorithm imposes a lower duty cycle resulting in moving the operating point to C, which is separate from the MPP in the new curve. On the other hand, when using our modified P&O technique, it is noted when Δ*P* > 0 and Δ*V* > 0, Δ*V* will be negative because the offset is positive (see the flowchart in Figure 3) causing the duty cycle to decrease and, consequently, the voltage to decrease as indicated in Figure 5, which shows the new operating point at D implying a faster response toward the new MPP. The previous discussion demonstrates that our modified P&O algorithm is a drift-free technique.

**Figure 5.** Drift analysis of conventional P&O versus modified P&O algorithms when considering a rapid increase in irradiance.

#### **4. Results and Discussion**

This section is concerned with presenting the tests of traditional and enhanced techniques: P&O, INC, and the modified P&O and INC, using MATLAB-Simulink to explore the steady-state and dynamic performance results by utilizing two distinct kinds of solar panels, MSX60 and ST40, whose types are polycrystalline thin films, respectively. Simulations are carried out and presented for standard test conditions and varying operating circumstances in ambient temperature and incident solar irradiation to investigate and compare the algorithm's efficiency regarding the response time desired to trace the peak power point and steady-state power oscillations and converter tracking efficiency.
