Enhanced leader particle swarm optimisation (ELPSO): An efficient algorithm for parameter estimation of photovoltaic (PV) cells and modules

Today, photovoltaic (PV) systems, are generating a significant share of electric power. Parameter estimation of photovoltaic cells and modules is a hot research topic and plays an important role in modelling PV systems. This problem is commonly converted into an optimisation problem and is solved by nietaheuristic optimisation algorithms. Among metaheuristic optimisation algorithms, particle swarm optimisation (PSO) is a popular leader based stochastic optimisation algorithm. However, premature convergence is the main drawback of PSO which does not let it to provide high-quality solutions in multimodal problems such as PV cells/modules parameter estimation. In PSO, all particles are pulled toward the leader, so the leader can significantly affect collective performance of the particles. A high-quality leader may pull all particles toward good regions of search space and vice versa. Therefore, in this research, an improved PSO variant, with enhanced leader, named as enhanced leader PSO (ELPSO) is used. In ELPSO, by enhancing the leader through a five-staged successive mutation strategy, the premature convergence problem is mitigated in a way that more accurate circuit model parameters are achieved in the PV cell/module parameter estimation problem. RTC France silicon solar cell, STM6.40/36 module with monocrystalline cells and PVM 752 GaAs thin film cell have been used as the case studies of this research. Parameter estimation results for various PV cells and modules of different technologies confirm that in most of the cases, ELPSO outperforms conventional PSO and a couple of other state of the art optimisation algorithms.