Photovoltaic array reconstruction method based on divide-and-conquer Q-learning algorithm

Abstract

According to the photovoltaic array reconstruction method based on the divide-and-conquer Q-learning algorithm; a large amount of historical data or environmental parameters are not needed, the number of peak values in a PV curve is reduced, the PV curve is smoother, mismatch loss is reduced, the array output power is improved, global optimum can be achieved, the application range is not limited, and the accuracy of the reconstruction process is high. An unsupervised training algorithm is adopted, an unknown environment is known through learning and experiences, and a large amount of historical data or environmental parameters are not needed. The integration of the divide-and-conquer principle further ensures the rapidity of the array reconstruction method; according to the photovoltaic array reconstruction method based on divide-and-conquer reinforcement learning, the performance of a photovoltaic array under array mismatch conditions such as local shading is rapidly improved, the number of peak values in a PV curve is reduced, the PV curve is smoother, mismatch loss is reduced, and array output power is improved. The accuracy of the reconstruction process is ensured by improving the training target, the stepping reward and the training termination criterion of the Q-learning algorithm.

Description

technical field [0001] The invention relates to the technical field of photovoltaic power generation, in particular to a photovoltaic array reconstruction method based on a divide-and-conquer Q-learning algorithm. Background technique [0002] Nowadays, the consumption of electric energy is gradually increasing all over the world, so the demand for power supply has also greatly increased. Traditional electricity comes from fossil fuels such as coal, natural gas, and oil. However, they are not only unsustainable energy sources, but also cause some unnecessary environmental impacts, such as global warming, rising sea levels, and harmful gas emissions. On the other hand, solar energy is a renewable and clean resource, and with the increasingly lower cost of photovoltaic (PV) panels, the improvement of power conversion efficiency, and the perfection of maximum power point tracking (MPPT) technology, photovoltaic power generation systems have become Widely established in China, ...

AI Technical Summary

Problems solved by technology

However, mathematically based reconstruction methods can only be used in some specific applications

In addition, the assumptions made in order to reduce the calculation time also reduce the accuracy of the method

Images