Recursive bit assignment with neural reference adaptive step (RNA) MPPT algorithm for photovoltaic system

Hegazy, Eman; Shokair, Mona; Saad, Waleed

doi:10.1038/s41598-023-28982-6

Cited by 2 publications

(1 citation statement)

References 30 publications

(35 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The approach proposed by the authors successfully improved the time it takes to track and reduced the amount of steady state oscillation near the MPP. Recursive bit assignment with neural reference adaptive step (RNA) algorithm proposed by the authors 17 are not able to track the global peak effectively. The introduction of ANN based hybrid MPPT methods draws more interest from the researchers 18 – 26 .…”

Section: Introductionmentioning

confidence: 99%

Performance optimization of interleaved boost converter with ANN supported adaptable stepped-scaled P&O based MPPT for solar powered applications

Krishnaram,

Padmanabhan,

Alsaif

et al. 2024

Sci Rep

View full text Add to dashboard Cite

Solar energy is the most promising among many renewable energy sources to meet the increasing demand. Photovoltaic (PV) based power generating solutions are expected to gain popularity as a power source for different applications, including independent and grid connected loads, due to their cleanliness, high performance, and high dependability. The efficacy of photovoltaic systems is impacted by several elements, including geographical location, positioning, shadowing effects, and local climate conditions. In order to fulfil the demands of loads, an interleaved boost converter is utilized, which has a reduced number of filters with less stress on the devices. Solar powered systems employ several maximum power point tracking (MPPT) methodologies. However, when there is partial shading, many power peaks arise, which complicates the identification of the overall peak. Although MPPT approaches are designed to measure and maintain the global maximum power point (GMPP), there are still significant oscillations observed around the GMPP with subpar settling time, tracking efficiency, and conversion efficiency. In this work, novel hybrid MPPT technique called artificial neural network supported adaptable stepped-scaled perturb and observe (ANN-ASSPO) method and water cycle optimization based perturb and observe (WCO-PO) have been proposed. Artificial neural network (ANN) has been used to determine the best scaling factor in ANN-ASSPO MPPT. Performance is enhanced in ANN-ASSPO MPPT by using the optimum scaling factor, particularly in situations when the irradiance is rapidly changing/partial shading conditions. Similarly, in WCO-PO MPPT water cycle optimization is used to determine the peak power when the PV panel is subjected to partial shading conditions. The performances of proposed hybrid MPPT ANN-ASSPO and WCO-PO techniques have been compared in terms of power generated, output voltage, average settling time and conversion efficiency. The MATLAB/Simulink tool is employed to carry out the experiment for this study.

show abstract

Section: Introductionmentioning

confidence: 99%

Performance optimization of interleaved boost converter with ANN supported adaptable stepped-scaled P&O based MPPT for solar powered applications

Krishnaram,

Padmanabhan,

Alsaif

et al. 2024

Sci Rep

View full text Add to dashboard Cite

show abstract

Leveraging opposition-based learning for solar photovoltaic model parameter estimation with exponential distribution optimization algorithm

Kullampalayam Murugaiyan,

Chandrasekaran,

Manoharan

et al. 2024

Sci Rep

View full text Add to dashboard Cite

Given the multi-model and nonlinear characteristics of photovoltaic (PV) models, parameter extraction presents a challenging problem. This challenge is exacerbated by the propensity of conventional algorithms to get trapped in local optima due to the complex nature of the problem. Accurate parameter estimation, nonetheless, is crucial due to its significant impact on the PV system’s performance, influencing both current and energy production. While traditional methods have provided reasonable results for PV model variables, they often require extensive computational resources, which impacts precision and robustness and results in many fitness evaluations. To address this problem, this paper presents an improved algorithm for PV parameter extraction, leveraging the opposition-based exponential distribution optimizer (OBEDO). The OBEDO method, equipped with opposition-based learning, provides an enhanced exploration capability and efficient exploitation of the search space, helping to mitigate the risk of entrapment in local optima. The proposed OBEDO algorithm is rigorously verified against state-of-the-art algorithms across various PV models, including single-diode, double-diode, three-diode, and photovoltaic module models. Practical and statistical results reveal that the OBEDO performs better than other algorithms in estimating parameters, demonstrating superior convergence speed, reliability, and accuracy. Moreover, the performance of the proposed algorithm is assessed using several case studies, further reinforcing its effectiveness. Therefore, the OBEDO, with its advantages in terms of computational efficiency and robustness, emerges as a promising solution for photovoltaic model parameter identification, making a significant contribution to enhancing the performance of PV systems.

show abstract

Recursive bit assignment with neural reference adaptive step (RNA) MPPT algorithm for photovoltaic system

Cited by 2 publications

References 30 publications

Performance optimization of interleaved boost converter with ANN supported adaptable stepped-scaled P&O based MPPT for solar powered applications

Performance optimization of interleaved boost converter with ANN supported adaptable stepped-scaled P&O based MPPT for solar powered applications

Leveraging opposition-based learning for solar photovoltaic model parameter estimation with exponential distribution optimization algorithm

Contact Info

Product

Resources

About