This paper proposes an enhanced sine-cosine algorithm (ESCA) to accurately identify parameters of photovoltaic (PV) cells, which are nonlinear and challenging to measure directly. ESCA introduces the concept of population average position to enhance exploration ability and incorporates a personal destination agent mutation mechanism and a competitive selection mechanism to improve search diversity and accuracy. The performance of ESCA is evaluated using single-diode, double-diode, three-diode, and photovoltaic module (PVM) models, comparing it with eight popular methods. Experimental results show that ESCA outperforms existing methods in terms of diversity maintenance, high efficiency, and stability, with smaller standard deviation statistics metrics for the three PVM models (PV-PWP201, STM6-40/36, and STP6-120/36). The proposed method is thus an accurate and reliable tool for parameter identification in PV cell models.This paper proposes an enhanced sine-cosine algorithm (ESCA) to accurately identify parameters of photovoltaic (PV) cells, which are nonlinear and challenging to measure directly. ESCA introduces the concept of population average position to enhance exploration ability and incorporates a personal destination agent mutation mechanism and a competitive selection mechanism to improve search diversity and accuracy. The performance of ESCA is evaluated using single-diode, double-diode, three-diode, and photovoltaic module (PVM) models, comparing it with eight popular methods. Experimental results show that ESCA outperforms existing methods in terms of diversity maintenance, high efficiency, and stability, with smaller standard deviation statistics metrics for the three PVM models (PV-PWP201, STM6-40/36, and STP6-120/36). The proposed method is thus an accurate and reliable tool for parameter identification in PV cell models.