Modeling of the operating characteristics of photovoltaic modules based on particle swarm optimization

This work evaluates the use of particle swarm optimization to extract the parameters that represent a photovoltaic cell in the model of one diode and five parameters (1D5P) to obtain current versus voltage characteristic curves (I x V) of a photovoltaic module operating under a wide range of temperature and irradiation scenarios. In all the modeling and simulation processes, only information of the module's data sheet provided by the manufacturers was used, without depending on results from other simulations or external data collections. To validate the study, the simulation of the current versus voltage (I x V) curves of a commercial photovoltaic module was made, and this was compared with data provided by the manufacturer and also compared with Chenni's method, which is consolidated by the literature. Errors of less than 0.3% were obtained for the simulations performed under the standard test conditions and in temperature and irradiation situations commonly found in practice, as is the case, of G = 700 W/m(2) and T-c = 40 degrees C, the 1D5P method showed an average error of approximately 1%, slightly surpassing the Chenni's method where the error was approximately 1.4%. Statement of Industrial Relevance In developing countries, such as Brazil, a significant increase in the portion referring to solar energy in the Brazilian energy grid is noted. As in 2016, this portion was only 0.02%, though in 2020, according to the Brazilian Solar Energy Association (ABSOLAR),(1) this portion had an increase of 75 times and reached the value of 1.5%. In addition, ABSOLAR's forecast for 2030 is that solar energy will reach 10% of the Brazilian energy grid. One of the reasons for this is that industrial consumer units are increasingly looking for alternative energy sources due to high energy inflation. Photovoltaic generators are an excellent option for industries because they are clean and sustainable energy.(2) It is silent and brings a financial return for the investor. Although, even with all the benefits, the photovoltaic generator has a high acquisition cost, especially for consumer units that demand much energy, as is the case of industries. Thus, correct and precise dimensioning for the photovoltaic generator is necessary. One way to increase the accuracy of dimensioning is to use simulation of the photovoltaic modules, which is the objective of this article. Novelty or Significance The developed method uses the particle swarm optimization with constriction factor to simulate I x V curves of a photovoltaic module for any temperature and irradiation values, using only the module's datasheet information. This study demonstrates that the proposed method presents more accurate results than consolidated methods in the literature, as is the case with the Chenni's method.(3)