[18a-PB01-1~25] 12.5 Organic and hybrid solar cells

[18a-PB01-8] A machine-learning study on the determination of equivalent circuit constants of solar cells from current – voltage characteristics on which noise signals are superimposed

Yuki Okuno¹, Takashi Kobayashi^2,3, Takashi Nagase^2,3, Kenjiro Fukuda⁴, 〇Hiroyoshi Naito^2,3 (1.Osaka Pref. Univ., 2.Osaka Metro. Univ., 3.RIMED, 4.RIKEN)

Keywords:organic solar cells, equivalent-circuit analysis, machine learning

Determination of the circuit constants of the one-diode equivalent circuit of a solar cell is important for solar-cell characterization. In this study, we developed a machine learning model that can analyze the current-voltage characteristics on which noise signal is superimposed. It is shown that this method can determine the equivalent circuit constants almost instantaneously (even when bagging is used, the equivalent-circuit constants can be determined from 1000 current-voltage characteristics in 4.2 s (CPU: Core i5, 1.80 GHz)).