Neural network-based adaptive global sliding mode MPPT controller design for stand-alone photovoltaic systems

Area of Science:

• Electrical Engineering
• Renewable Energy Systems
• Control Theory

Background:

• Growing energy demands and environmental concerns necessitate efficient renewable energy solutions.
• Photovoltaic (PV) systems are crucial but their energy output fluctuates with environmental factors like temperature and solar radiation.
• Effective Maximum Power Point Tracking (MPPT) is vital for optimizing energy yield from PV arrays under varying conditions.

Purpose of the Study:

• To present a nonlinear generalized global sliding mode controller (GGSMC) for enhanced maximum power extraction from PV arrays.
• To utilize a feed-forward neural network (FFNN) for generating optimal reference voltage under dynamic environmental conditions.
• To validate the proposed GGSMC strategy's performance against a standard nonlinear backstepping controller.

Main Methods:

• Design of a GGSMC controller integrated with a DC-DC buck-boost converter for PV energy harvesting.
• Implementation of a feed-forward neural network (FFNN) to provide dynamic reference voltage tracking.
• Modification of the sliding mode control to eliminate the reaching phase, ensuring continuous sliding mode operation.
• Simulation analysis using MATLAB/Simulink to evaluate system response, accuracy, and tracking speed.

Main Results:

• The proposed GGSMC strategy effectively tracks the FFNN-generated reference voltage under fluctuating temperature and sunlight.
• The control system demonstrated no chattering or harmonic distortions in its response.
• Simulation results confirmed the effectiveness, accuracy, and rapid tracking capabilities of the GGSMC.
• Comparative analysis showed superior or comparable performance against the nonlinear backstepping controller under abrupt environmental changes.

Conclusions:

• The developed GGSMC offers a robust and efficient solution for maximizing power output from PV systems.
• The elimination of the reaching phase and absence of chattering contribute to a stable and reliable control system.
• The FFNN-enhanced GGSMC is a promising strategy for optimizing PV energy generation in diverse environmental conditions.