Desenvolvimento de um método de monitoramento e detecção de defeitos em células fotovoltaicas utilizando termografia ativa modulada

Abstract

The occurrence of defects in photovoltaic cells is intrinsically related to a reduction in the efficiency and reliability of these devices. Therefore, monitoring techniques, such as lock-in thermography, electroluminescence, and I-V characteristic curve, are adopted to assess the integrity of the cells. In this study, a novel experimental procedure using lock-in thermography in photovoltaic cells is proposed aiming to improve the defect detection capability of the thermographic test. Conventional techniques that use pulse width modulation to stimulate photovoltaic cells usually operate at a fixed point on the I-V curve. Instead, a methodology based on a sinusoidal electrical voltage signal excitation is proposed, which expands the range of operational points of the cell during the test. Several traditional image processing techniques (principal component analysis, fast Fourier transform, four-step phase-shifting method, and two-dimensional Gaussian function method) were used alone to analyze the thermal images of defective and non-defective photovoltaic cells during the steady-state operation mode. The results obtained showed that this new approach provides a smoother and clearer delimitation of defective regions, enabling an adequate characterization of existing defects. Additionally, the phase images gained prominence in the characterization of defects, in particular for the frequencies of 50 mHz and 100 mHz, due to the contrast they presented with significant changes between defective and non-defective regions for different modulation frequencies and types of defects.