IEEE Journal of Photovoltaics · 2016 · 57 citations · 26 references
EngineeringElectrical EngineeringEnergy ManagementOptimizationPower ElectronicsEnergy Management SystemsPower ManagementPower OptimizationSolar Energy SystemsSolar Photovoltaic SystemsVoltage RegulationMaximum Power Point TrackingMaximum Power Point Tracking AlgorithmsOptimal Power ControlParticle Swarm ControlParticle Swarm OptimizationPhotovoltaic System OptimizationPower OscillationsVoltage
Nonuniform irradiation conditions (NUIC) of a photovoltaic (PV) array pose a major challenge to optimal power utilization due to multiple power peaks (consisting of multiple local peaks and one global peak) in power-voltage (P-V) characteristics. This paper proposes an adaptive velocity particle swarm optimization (AVPSO) algorithm for tracking the global maximum of the multiple peak P-V characteristics. The AVPSO algorithm continuously adjusts individual particle's weight factor and cognitive acceleration coefficient, depending upon its distance from the global best position during the tracking process. The advantage of the adaptive weight factor is reduced power oscillations in the region of global best position, while adaptive cognitive factor prevents the particles from getting trapped in local minima. Thus, the adaptive nature of the particle's velocity improves the global maximum power point tracking (GMPPT) time and power yield. Another feature involves limiting the particle's velocity to avoid skipping intermediate peaks. Additionally, AVPSO also includes the ability to sort particle's positions in each iteration, which prevents large changes in the PV voltage, thereby reducing the control effort of a PV voltage controller. The effectiveness of the proposed GMPPT algorithm is investigated for static and dynamic performance under different NUIC patterns and mitigation of local minima trapping issues through analysis, simulations, and extensive hardware experiments.
26