Maximum power point tracking for photovoltaic applications by using improved PSO under partial shading conditions

Kamarudin, Nornaim (2024) Maximum power point tracking for photovoltaic applications by using improved PSO under partial shading conditions. Masters thesis, Universiti Teknologi MARA, Shah Alam.

Abstract

Maximum power point tracking (MPPT) for photovoltaic (PV) systems is difficult, especially under partial shadowing. Traditional particle swarm optimisation (PSO) approaches have many drawbacks, such as growing particle populations, which slow the search for the global peak and lower PV system efficiency and stability. High inertia weight values can make particles move aggressively, exceeding the best solution and slowing convergence. Due to particle mobility reduction, low inertia weight values might trap particles in local minima prematurely. Incorrect acceleration coefficients can produce rapid convergence at a local optimum, preventing enticing investigation or slowing optimal solution movement, resulting in poor performance. The challenges can be overcome by developing a new hybrid inertia weight (HIW) for rapid tracking of the maximum power point (MPP) under different irradiance levels and partial shading conditions, enhancing the acceleration coefficient using a cosine method for accurate MPP tracking, and evaluating the improved PSO's accuracy, efficiency, and speed. A comprehensive research overview, PV system modelling under PSC, buck-boost converter design for PV applications, and PSO algorithm refinement by optimising initialisation conditions, population size, acceleration coefficients, and hybrid inertia weight are included. Simulink implements the updated PSO algorithm for rigorous performance evaluation. Results show significant gains using proposed strategies. An optimal population size of 5 has a 96.28% efficiency average. The Naim 1 acceleration coefficient approach had 90.98% efficiency, but the cosine-based method performed better. The HIW method surpassed prior methods with 97.38% efficiency, 100% accuracy, and 1.7899 seconds convergence. Further dynamic condition research showed that the enhanced PSO consistently outperformed conventional, LQR, and ED PSO. The HIW PSO had an average efficiency of 97.1% and accuracy of 100% under constant irradiance, PSC, and a combination of both. For constant irradiance, PSC, and a mix of both, HIW averaged 1.4020 s, 1.3 s, and 1.1196 s for convergence speeds. The proposed HIW improved PSO convergence, accuracy, and efficiency in dynamic situations. These advances are essential for more reliable and efficient PV systems, especially in partial shadow. The improved PSO technique improves PV system power production reliability and efficiency and boosts public acceptance for renewable energy sources by fixing their most fundamental flaws.

Metadata

Item Type: Thesis (Masters)
Creators:
Creators
Email / ID Num.
Kamarudin, Nornaim
UNSPECIFIED
Contributors:
Contribution
Name
Email / ID Num.
Thesis advisor
Abd Samat, Ahmad Asri
UNSPECIFIED
Thesis advisor
Osman, Khusairi
UNSPECIFIED
Thesis advisor
Tajuddin, Mohammad Faridun Naim
UNSPECIFIED
Subjects: Q Science > Q Science (General) > Back propagation (Artificial intelligence)
T Technology > TK Electrical engineering. Electronics. Nuclear engineering > Production of electric energy or power
Divisions: Universiti Teknologi MARA, Shah Alam > College of Engineering
Programme: Master of Science (Electrical Engineering)
Date: November 2024
URI: https://ir.uitm.edu.my/id/eprint/144027
Edit Item
Edit Item

Download

[thumbnail of 144027.pdf] Text
144027.pdf

Download (115kB)

Digital Copy

Digital (fulltext) is available at:

Physical Copy

Physical status and holdings:
Item Status:

ID Number

144027

Indexing

Statistic

Statistic details