Numerical Modeling and Neural Network Optimization for Advanced Solar Panel Efficiency DOI
Udit Mamodiya, Indra Kishor, Mohammed Amin Almaiah

et al.

Research Square (Research Square), Journal Year: 2025, Volume and Issue: unknown

Published: April 16, 2025

Abstract Maximizing output from renewable solar panels requires higher efficiency. Conventionally, such optimization techniques - MPPT (Maximum Power Point Tracking) along with heuristic algorithms suffer significantly slow adaptability and track sub optimality under dynamic environments. This article proposes a numerical modeling framework hybrid AI models, combining physics-informed neural networks RL for real-time of orientation in panels. The methodology uses precise energy transformation analysis, deep learning-based dynamically adjusts the angles to maximize power output. A self-learning adaptive network is developed improve tracking accuracy based on irradiance temperature variations. Moreover, an Edge architecture introduced make low-latency decisions reduced dependency cloud computation, thus improving efficiency system. Besides, advanced model CNN-LSTM applied forecasting predictive control maximum yield. Experimental validation was performed using UTL 335W 330W PV modules, where data acquisition followed by AI-driven optimization. Results show increase yield 10–15% compared traditional systems, while computations are 40–50% faster AI-based modeling. proposed approach achieves 25% lower error (RMSE/MAE) 30% consumption through implementation. study sets up new paradigm AI-integrated optimization, which ensures enhanced performance practical deployment. findings advance intelligent set benchmark management.

Language: Английский

A high-speed MPPT based horse herd optimization algorithm with dynamic linear active disturbance rejection control for PV battery charging system DOI Creative Commons

AL-Wesabi Ibrahim,

Jiazhu Xu, Imad Aboudrar

et al.

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Jan. 25, 2025

This study first proposes an innovative method for optimizing the maximum power extraction from photovoltaic (PV) systems during dynamic and static environmental conditions (DSEC) by applying horse herd optimization algorithm (HHOA). The HHOA is a bio-inspired technique that mimics motion cycles of entire horses. Next, linear active disturbance rejection control (LADRC) was applied to monitor HHOA's reference voltage output. LADRC, known managing uncertainties disturbances, improves anti-interference capacity point tracking (MPPT) speeds up system's response rate. Then, in comparison traditional (perturb & observe; P&O) metaheuristic algorithms (conventional particle swarm optimization; CPSO, grasshopper GHO, deterministic PSO; DPSO) through DSEC, simulations results demonstrate combination HHOA-LADRC can successfully track global peak (GMP) with less fluctuations quicker convergence time. Finally, experimental investigation proposed accomplished NI PXIE-1071 Hardware-In-Loop (HIL) prototype. output findings show effectiveness provided may approach value higher than 99%, showed rate converging oscillations detection mechanism.

Language: Английский

Citations

1
Optimized Energy Management Strategy for an Autonomous DC Microgrid Integrating PV/Wind/Battery/Diesel-Based Hybrid PSO-GA-LADRC Through SAPF DOI Creative Commons

AL-Wesabi Ibrahim,

Jiazhu Xu, Abdullrahman A. Al-Shamma’a

et al.

Technologies, Journal Year: 2024, Volume and Issue: 12(11), P. 226 - 226

Published: Nov. 11, 2024

This study focuses on microgrid systems incorporating hybrid renewable energy sources (HRESs) with battery storage (BES), both essential for ensuring reliable and consistent operation in off-grid standalone systems. The proposed system includes solar energy, a wind source synchronous turbine, BES. Hybrid particle swarm optimizer (PSO) genetic algorithm (GA) combined active disturbance rejection control (ADRC) (PSO-GA-ADRC) are developed to regulate the frequency amplitude of AC bus voltage via load-side converter (LSC) under various operating conditions. approach further enables efficient management accessible generation general consumption through bidirectional battery-side (BSC). Additionally, method also enhances power quality across link mentoring photovoltaic (PV) inverter function as shunt filter (SAPF), providing desired harmonic-current element nonlinear local loads well. Equipped an extended state observer (ESO), PSO-GA-ADRC provides estimation compensation disturbances such modeling errors parameter fluctuations, stable solution interior current loops. positive results from hardware-in-the-loop (HIL) experimental confirm effectiveness robustness this strategy maintaining real-world scenarios.

Language: Английский

Citations

3
An efficient linear-extrapolation catch-fish algorithm for maximizing the harvested power from thermoelectric generators sources DOI

AL-Wesabi Ibrahim,

Jiazhu Xu, Hassan M. Hussein Farh

et al.

Applied Thermal Engineering, Journal Year: 2025, Volume and Issue: unknown, P. 125916 - 125916

Published: Feb. 1, 2025

Language: Английский

Citations

0
Numerical Modeling and Neural Network Optimization for Advanced Solar Panel Efficiency DOI
Udit Mamodiya, Indra Kishor, Mohammed Amin Almaiah

et al.

Research Square (Research Square), Journal Year: 2025, Volume and Issue: unknown

Published: April 16, 2025

Abstract Maximizing output from renewable solar panels requires higher efficiency. Conventionally, such optimization techniques - MPPT (Maximum Power Point Tracking) along with heuristic algorithms suffer significantly slow adaptability and track sub optimality under dynamic environments. This article proposes a numerical modeling framework hybrid AI models, combining physics-informed neural networks RL for real-time of orientation in panels. The methodology uses precise energy transformation analysis, deep learning-based dynamically adjusts the angles to maximize power output. A self-learning adaptive network is developed improve tracking accuracy based on irradiance temperature variations. Moreover, an Edge architecture introduced make low-latency decisions reduced dependency cloud computation, thus improving efficiency system. Besides, advanced model CNN-LSTM applied forecasting predictive control maximum yield. Experimental validation was performed using UTL 335W 330W PV modules, where data acquisition followed by AI-driven optimization. Results show increase yield 10–15% compared traditional systems, while computations are 40–50% faster AI-based modeling. proposed approach achieves 25% lower error (RMSE/MAE) 30% consumption through implementation. study sets up new paradigm AI-integrated optimization, which ensures enhanced performance practical deployment. findings advance intelligent set benchmark management.

Language: Английский

Citations

0