top of page

MATLAB Implementation of PSO Trained ANFIS MPPT for Solar PV System

Writer's picture: lms editorlms editor

MATLAB Implementation of PSO Trained ANFIS MPPT for Solar PV System

Understanding ANFIS Structure

  • Layers of ANFIS: The ANFIS model comprises five layers, each serving a specific function. Layers one and four are tunable, with parameters that need adjustment for optimal performance.

  • Tunable Parameters: In layer one, we distribute inputs using membership functions, while layer four involves calculations that utilize tunable parameters such as P1P1P1, Q1Q1Q1, and R1R1R1.

  • Hybrid Algorithms: The tunable parameters are optimized using a hybrid algorithm that combines backpropagation and least squares methods.

Simulation Process

Data Generation

  • Input Parameters: The input data for the ANFIS model is generated based on two primary parameters: temperature and solar irradiation.

  • Output Target: The output target for the model is the maximum power voltage (VMP), derived from the solar PV model.

Training the ANFIS Model

  • Training Program: The training process involves loading the generated data and using PSO to tune the parameters of the ANFIS model.

  • Cost Function: A root mean square error is calculated to evaluate the performance, with the objective of minimizing this error throughout the training iterations.

Implementation of MPPT Algorithm

Model Development

  • PV Panel Specifications: The model uses a 250W PV panel with defined operational parameters at standard test conditions.

  • Load and Boost Converter: A resistive load is connected to the PV panel through a boost converter, which is controlled by the MPPT algorithm.

System Response

  • Load Variation: The system is tested under varying load conditions to observe its response in terms of power, voltage, and current.

  • Radiation Changes: The model is further tested by changing the solar irradiation levels to verify that maximum power extraction is maintained under different conditions.

Results and Performance

  • Efficiency of MPPT: Throughout the simulations, the PSO-trained ANFIS consistently extracts maximum power from the solar PV panel, demonstrating the effectiveness of the approach.

  • Data Visualization: The simulation results are visualized through various plots that depict the relationship between training and test data, showcasing the model's accuracy.

Conclusion

This implementation of a PSO-trained ANFIS MPPT for solar PV systems illustrates a powerful method for optimizing solar energy extraction. The successful tuning of parameters and the system's adaptability to changing conditions highlight its potential for real-world applications.

15 views0 comments

Comments


bottom of page