MESSAI.iO

AI summary

65% confidence

This study applied three types of particle swarm optimization algorithms to extract parameters from proton exchange membrane fuel cells and photovoltaic cells, with momentum PSO yielding the most accurate results. The use of momentum PSO enabled rapid convergence and stable parameter extraction. The findings of this study have implications for the optimization and modeling of complex energy systems.

Generated by MESSAI extraction pipeline · review against source PDF

Extraction

Reported parameters

1 extracted value

View extracted values

No 3D model is mapped to this paper yet. Parameter ranges above still place reported values on the literature distribution.

Abstract

As carriers of green energy, proton exchange membrane fuel cells (PEMFCs) and photovoltaic (PV) cells are complex and nonlinear multivariate systems. For simulation analysis, optimization control, efficacy prediction, and fault diagnosis, it is crucial to rapidly and accurately establish reliability modules and extract parameters from the system modules. This study employed three types of particle swarm optimization (PSO) algorithms to find the optimal parameters of two energy models by minimizing the sum squared errors (SSE) and roots mean squared errors (RMSE). The three algorithms are inertia weight PSO, constriction PSO, and momentum PSO. The obtained calculation results of these three algorithms were compared with those obtained using algorithms from other relevant studies. This study revealed that the use of momentum PSO enables rapid convergence (under 30 convergence times) and the most accurate modeling and yields the most stable parameter extraction (SSE of PEMFC is 2.0656, RMSE of PV cells is 8.839 · 10⁻⁴). In summary, momentum PSO is the algorithm that is most suitable for system parameter identification with multiple dimensions and complex modules.

Key findings

  • Momentum PSO achieved the most accurate modeling and parameter extraction
  • Momentum PSO enabled rapid convergence in under 30 convergence times
  • The sum squared error for PEMFC was 2.0656 and the root mean squared error for PV cells was 8.839 · 10⁻⁴

Keywords

Particle swarm optimizationPhotovoltaic systemAlgorithmConvergence (economics)Proton exchange membrane fuel cellMean squared error

Identifiers

DOI
10.3390/en14030619
Journal
Energies
Year
2021