A Stand-Alone PV-PEMFC System based SM-ANN-MPPT Controller: Solar Pumping Application using PMSM

Maximum Power Point Tracking (MPPT) is a tool to optimize the use of the produced power in renewable energy systems, particularly in photovoltaic (PV) systems. Many algorithms have been developed and reviewed in the literature. In this work, a Sliding Mode-Artificial Neural Network-based MPPT (SM-ANN-MPPT) technique is proved. A comparison is made with one of the simplest MPPTs in the literature Perturb and Observe (P&O). It has been found that the minimization of the PV power ripples and the dissipated energies have been gained using the SM-ANN-MPPT controller. The proposed controller is applied for a standalone PV system with a Proton Exchange Membrane Fuel Cell (PVPEMFC) considering variable irradiation and temperature. The application of the system is dedicated to a solar pumping system using a three-phase Permanent Magnet Synchronous Motor (PMSM). The control of the PMSM is ensured by a Direct Torque Control-Space Vector Modulation (DTC-SVM) approach. Two cases are tested: with imposed speed profile, and with a solar daily profile. Two cases of load profiles are also treated: constant and variable loads. The control behaviour is simulated and the results are revealed justifying the feats of the DTC-SVM PMSM drive. Results show improvement in robustness and stability under real daily climatic conditions.