摘要
The development and application of photovoltaic (PV) systems are becoming increasingly more important as the global need for energy resources expands and environmental protection becomes more highly valued. Parameters of PV models can be identified by measuring their current-voltage (I-V) characteristic curves. Identifying these parameters quickly, accurately and reliably is critical in determining the operating status of in situ PV arrays and, in turn, optimising solar energy conversions. To achieve both fast and accurate parameter identification with high reliability, a new algorithm called algorithm based on SSO and Nelder-Mead simplex (NMS) (SSSO) based on the simplified swarm optimisation (SSO) and the NMS is proposed in this study. To demonstrate the performance of SSSO in identifying solar cell system parameters, its performance on the single diode model and the double diode model was compared with existing algorithms in terms of both fitness value and run time. The experiment results indicate that SSSO outperformed the compared algorithms in both run time and standard deviation of fitness value. © The Institution of Engineering and Technology 2017.