Image-based Power Point Tracking for Photovoltaic (PV) Systems

A Photovoltaic (PV) module is comprised of a number of solar cells. The efficiency of a PV module depends on two factors: the amount of incident sunlight, and the electrical characteristics of the appliance that will be drawing power from the PV module. To ensure efficient power transfer to the appliance (“load”), it’s imperative that the electrical characteristics of the load are tuned to the output from the solar cells. Under homogeneous lighting conditions, there’s a single peak in the PV module’s output power curves, and electrical characteristics of any load can be locked onto that singular peak. However, under partial shading conditions, multiple peaks show up in the output power curves. Losses occur when the electrical characteristics are tuned to a local power peak instead of the global power peak which would guarantee maximum power transfer efficiency. Maximum Power Point Tracking (MPPT) techniques are a class of algorithms that aim to lock on to the global power peak out of the multiple local power peaks.

Current MPPT methods find the global peak by periodically scanning the power curves, an approach that is inefficient because irradiances change between two successive scans. Therefore, the global peak being accurately found is not guaranteed. Instead of performing periodic scans, a novel analytic technique is proposed that makes use of an optical camera. Images of a PV module are continuously captured to obtain the reflected irradiance in the direction of the camera. This data is used to compute the incident irradiance levels received by the solar cells, which is then fed into a mathematical model of the PV module to determine the global peak in the power curves analytically. Using a camera to produce data for an analytic solution is more efficient in correctly identifying the global peak than performing periodic scans of the output power curves.