Renewable energy sources are becoming increasingly critical for combating global electricity shortages and mitigating environmental pollution. Among these, solar energy stands out for it s abundance, minimal environmental impact, and widespread availability across diverse regions. However, optimizing solar energy production is essential to maximize its potential by reducing costs, improving efficiency, and fully utilizing this ever -present resource. This research introduces two novel, AI -powered nonlinear control methodologies designed for precise Maximum Power Point Tracking (MPPT) in photovoltaic (PV) systems. These methodologies are specifically crafted to handle dynamic changes in solar irradiance and temperature with exceptional adaptability. The proposed techniques' performance is rigorously evaluated through simulations conducted within the MATLAB/Simulink environment. The evaluation focuses on their effectiveness under various system conditions, employing two distinct DC -DC converter configurations for a comprehensive analysis.

Photovoltaic energy is a very important renewable energy source. However, modern energy strategies are increasingly integrating the PV system into the grid. In the PV grid connected, energy improvement is a continuous process that includes maximizing PV power, regulating DC link voltage, compensating reactive power, and minimizing total harmonic distortion. In this paper, the two-stage configuration of the transformerless system is adopted for PV grid connections. In this configuration, it is aimed at improving the DC link voltage controller. Three controls can be compared, such as the conventional DC Link voltage controller, the back-stepping voltage controller, and the proposed DC Link voltage controller. These controls are tested in MATLAB/simulink in order to evaluate the performance of each one. The results show some advantage for the proposed control.

In this paper, we present a comparison of three types of algorithms: Perturb and Observe (P&O), Incremental Conductance (InCnd), and Fuzzy Logic Controller (FLC) that track the maximum power point (MPP) of a photovoltaic system (PV) over varying conditions of solar irradiation and temperature. The PV system is composed of solar panels, a resistive load, and an MPPT controller with pulse width modulation (PWM) technique for driving the DC-DC Quadratic Boost converter (QBC). This comparison, based on three criteria, namely stability, time response and, power efficiency, demonstrates the benefit of employing an MPPT with variable step monitoring. The energy obtained using those three algorithms is practically similar, with a considerable improvement for the Fuzzy Logic Controller.

As PV energy is clean and renewable, more and more PVG is connected to the grid. In this paper, PV grid connected is presented in a two-stage configuration. The first stage is a DC bus. It is composed by a BOOST converter with its MPPT control. This control is based on Perturb and observe method. This method is used under three scenarios, conventional method, optimized method by two PI controllers and modified method. The second stage is the AC bus. It is composed of a VSI inverter and LCL filter with its control, where there are three loops are figured such as outer current, voltage and inner current. The reactive power compensation and THD compensation are considered in order to improve energy flow between PVG and the three phase grid. Parameters PI controller are optimized by an easy analytical method. The theoretical study is tested and verified by simulation under MATLAB.Simulink.

In this paper, an identification problem for nonlinear models is explored and an improved fuzzy identification method based on the heterogeneous Multi-swarm PSO (MsPSO) algorithm is proposed in order to obtain an optimal T-S fuzzy model. However, this simple homogeneous search behavior is not always optimal to find the potential solution to a special problem, and it may trap the individuals into local regions leading to premature convergence. To improve the performance of MsPSO, the particles should be able to adaptively changing their original trajectories to explore new search space. In fact, a new multiswarm particle swarm optimization algorithm using an adaptive inertia weight, denoted AIMsPSO, has been presented in order to improve the performance of constructing the T-S fuzzy system.