Fast and Accurate Maximum Power Point Tracking for Concentrating Photovoltaic Systems

Background

Maximum power point tracking (MPPT) is a method for power optimization in photovoltaic (PV) systems using a highly efficient DC-to-DC converter that adjusts to changes in atmospheric conditions, shading of PV modules or cells, etc. At present, the standard method for regulating MPPT is the “perturb and observe” (P&O) algorithm, also known as the hill-climbing algorithm, where perturbation of a reference voltage is controlled by a feedback loop in response to the power output. While relatively simple, the P&O algorithm does impose a tradeoff in dynamic range and tracking time, meaning that P&O-regulated voltages have a tendency to drift away from the maximum power point under rapidly changing environmental conditions, or can oscillate around the maximum power point. Other algorithms have been tried, but they too have various limitations in terms of residual drift or are dependent on various ad hoc adjustments that are not generally applicable. Thus, a remaining challenge for MPPT is to develop algorithms that have a less severe trade-off between tracking speed, tracking accuracy, and customization for particular PV cells.

Description

Researchers at the University of California, Merced (UCM) have invented an improved MPPT system based on an unbounded binary search algorithm for optimizing each PV cell. The UCM MPPT system does not require ad hoc parameter adjustments in order to track at a rate that is a logarithmic function of the voltage search range.

Applications

The UCM MPPT system may find wide application in optimizing the power output of concentrating PV (CPV) cells or PV cells with a single local maximum power point.

Advantages

The UCM MPPT system is faster and more accurate than previous MPPT systems, making it more suitable for optimizing the power output of CPV units that are subjected to rapid environmental changes due to such factors as partial shading or self-shading and changes in atmospheric conditions. It achieves a fast tracking speed that is a logarithmic function of the tracking space without requiring ad hoc parameters in the algorithm being customized for particular CPV cells.

Inventor

Yong Sin Kim
Roland Winston