Comparative Study Between the Different MPPT Techniques

This paper proposes a new hybrid controller based MPPT. This one is composed of the Artificial Neural Network and Integral Feedback Linearization Controller (ANN-IFLC). The ANN is used to produce the reference of optimal voltage that corresponds to the maximal power. While the IFLC is designed to track the voltage reference. The integral action is added to ensure zero-static error. Then, a comparison between different existing Maximum Power Point Tracking (MPPT) techniques, which are divided between the classical algorithms and the hybrid controllers, is made. The classical algorithms, treated in this study, are the Perturb and Observe (P & O) and Incremental Conductance (INC). While the hybrid controllers are INC-BSC, P & O-BSC, ANN-ISMC, ANN-BSC, and ANN-IFLC. As can be noticed, these controllers present the combination between the nonlinear controllers and the Artificial Neural network (ANN) or the classical algorithms (P & O and INC). Effectively, the nonlinear controllers used are the Sliding Mode Controller (SMC), the Integral Sliding Mode Controller (ISMC), the Backstepping Controller (BSC) and the proposed IFL Controller. The proposed photovoltaic system consists of the photovoltaic module (Reference: S6M2G240), of the Boost converter and the resistive load. The MPPT techniques are tested using Matlab software. The results show that the hybrid controllers characterize by the tracking performances better than the classical methods. Moreover, the Artificial Neural Network predicts quickly and accurately the Maximum Power Point (MPP) under uniform meteorological conditions. Also, the PV voltage produced, using IFLC or ISMC, has fewer oscillations around its optimum thanks to the integral action added.