An Improved Model Combining Evolutionary Algorithm and Neural Networks for PV Maximum Power Point Tracking

• Published in: IEEE access Vol. 7; pp. 2823 - 2827
• Main Authors: Wang, Haixin, Shen, Jianxin
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 2019; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Artificial neural networks; Back propagation; Computer simulation; Control algorithms; Control methods; Control simulation; Control theory; Electric potential; Evolutionary algorithms; Evolutionary computation; Genetic algorithms; Mathematical model; Maximum power tracking; mind evolutionary algorithm; Neural networks; Optimization; Photovoltaic cells; photovoltaic maximum power tracking; Predictive control; Simulation; Training; Unmanned aerial vehicles; Voltage; Voltage control
• ISSN: 2169-3536; 2169-3536
• DOI: 10.1109/ACCESS.2018.2881888

Aiming at the large error of the traditional constant control method in predicting the maximum power of solar UAV, this paper proposed an improved mind evolutionary algorithm combined with BP neural network (BPNN), in which the improved mind evolutionary algorithm optimizes the BPNN. The optimized model is used to predict the voltage at the maximum power of the panel in the UAV. The constant voltage parameter based on the conventional constant pressure control algorithm is replaced by this value. At the same time, a new control simulation model of constant voltage solar panel maximum power tracking based on the improved mind evolution algorithm optimize BPNN was built. At last, the algorithm was simulated and validated in the MATLAB/Simulink environment. The simulation results show that this algorithm is better than simply using BPNN and genetic algorithm optimize BPNN stability and higher accuracy.