Online identification of photovoltaic source parameters by using a genetic algorithm

In this paper, an efficient method for the online identification of the photovoltaic single-diode model parameters is proposed. The combination of a genetic algorithm with explicit equations allows obtaining precise results without the direct measurement of short circuit current and open circuit voltage that is typically used in offline identification methods. Since the proposed method requires only voltage and current values close to the maximum power point, it can be easily integrated into any photovoltaic system, and it operates online without compromising the power production. The proposed approach has been implemented and tested on an embedded system, and it exhibits a good performance for monitoring/diagnosis applications.