Otimização de sistemas fotovoltaicos.

Abstract

The distributed energy generation model has frequently been focused, showing advantages over the centralized generation model. Renewable energy sources such as wind and photovoltaics (PV) have a key role in these systems, having low environmental impact. In this work PV systems will be focused. Photovoltaic systems are usually built of a set of PV generator modules, batteries, charge controller and inverter, having several optimization opportunities in these components. One possible optimization is employing of maximum power point tracking (MPPT) methods, which will change the generator's operating point in such a way as to maximize the generated power. In this work a new MPPT method is proposed, employing the generator's nameplate data for accelerated tracking. A DC-DC converter capable of executing the method is also developed. The developed MPPT method is based on the equivalent circuit model of the PV generator, and requires no previous knowledge of the load, no expensive sensors, and no excessive processing power. The model is determined from nameplate data, and corrected using the operating temperature. From the model a special set of curves is calculated, describing all possible locations of the MPP on the voltage-current plane. These curves are named MPP Locus, and are employed to determine the optimum working voltage for the generator. Partial shading conditions are considered, and corrected by fusing the model-based method with a perturb-and-observe type. With this method, a tracking speed gain of up to 70:1 is achieved when compared to the non-model-based methods. The realtime computational complexity is reduced by employing a two-dimensional lookup table with 512 points and bilinear interpolation. Even thought the calculation of the table is processing intensive, this is required only once and not necessarily on the target processor, therefore achieving satisfactory results even on systems with limited processing power. The DC-DC converter studied is a current-fed push-pull. This converter is specially adequate for PV systems, where th input voltage is usually low (10.5-60V) and a high gain is required. Galvanic isolation is also provided in order to protect the user from accidental shock in the low-voltage circuit. A control law is developed in order to enable the use of the MPPT method with the converter. A dedicated control loop is used to prevent core saturation, excluding the related losses, and allowing a smaller core size. The control law is configured in such a way that MPPT is enabled only if the load can sink all the power, and disabled otherwise, removing the need for a minimum load guarantee. Experimental results demonstrate the converter's functionality, and simulation results are used to validate the control law.