Authors : Razieh Khanaki, Mohd Amran Mohd Radzi, Shahrooz Hajighorbani and Mohammad Hamiruce Marhaban
Abstract: This study presents an improved Maximum Power Point Tracking (MPPT) controller using Artificial Neural Network (ANN) which is evaluated under different condition of solar irradiance and cell temperature. This intelligent method is compared with Perturbation and Observation (P&O) method which is the most popular and commonly used conventional MPPT controller. The transient and steady state responses are presented and compared for both high and low solar irradiations as well as the dynamic responses. The control system is implemented on eZdsp TMF28335 Digital Signal Processor (DSP). Experimental results are provided for both high and low irradiations, at the same condition of cell temperature and solar irradiation applied in simulation work. The results show that ANN MPPT has smaller tracking time and provides higher efficiency than P&O with different step-sizes, under both high and low solar irradiations. In addition, in term of dynamic responses, the ANN MPPT controller is much faster than P&O MPPT at locating and tracking the Maximum Power Point (MPP), in case of changing solar irradiation condition.
Razieh Khanaki, Mohd Amran Mohd Radzi, Shahrooz Hajighorbani and Mohammad Hamiruce Marhaban, 2016. Evaluation of Improved MPPT-Based ANN Controller for PV Standalone System. Journal of Engineering and Applied Sciences, 11: 1972-1980.