Estimation-Based Maximum Power Point Tracking in a Self-Balancing Photovoltaic Battery Energy Storage System

Abstract

This paper examines the problem of controlling the exchange of current in photovoltaic-plus-storage systems to achieve photovoltaic (PV) maximum power point tracking (MPPT). This work is motivated by the need for MPPT algorithms that are less costly and complex to implement in PV farms with integrated battery energy storage. We study the online optimal control of a “hybrid” PV/lithium (Li)-ion battery integration topology that is self-balancing in nature. The self-balancing behavior ensures that the state of charge (SOC) across different cells balances to the same stable equilibrium value without needing any balancing power electronics, thereby significantly reducing the integration cost. The DC–DC converters in this hybrid system are controlled to achieve PV MPPT that maximizes energy generation and storage. However, sensing needs for traditional MPPT controllers can render the hybrid system unnecessarily complex and costly. We surmount this problem by: (i) developing a novel model-based PV power estimation algorithm that only requires voltage measurement, and (ii) using this algorithm together with extremum-seeking (ES) control to achieve closed-loop, estimation-based PV MPPT. Simulation case studies show that this estimation-based MPPT controller is able to harness more than 99% of the maximum available solar energy under different irradiation profiles.