Multi-peak-value MPPT algorithm

Abstract

The invention discloses a multi-peak-value MPPT algorithm. According to the algorithm, during starting, a large-step-length global scanning algorithm is utilized to quickly locate all peak value intervals, afterwards, a filtering algorithm is utilized to reduce the peak value intervals, dichotomy is utilized to find local peak values of all the filtered 1-5 peak value intervals, and finally the maximum value of the local peak values are regarded as the global maximum power point; through an acknowledgment mechanism, a solved GMPP is confirmed, the algorithm is prevented from being misjudged, and finally a three-point method algorithm is utilized to make the system dynamically track the GMPP under a changing environment. By means of the multi-peak-value MPPT algorithm, the photovoltaic system maximum power point tracing speed is increased, so that the system quickly adapts to the environment change, the algorithm is prevented from being involved into the local optimal point, the misjudgment rate of the algorithm is reduced, the algorithm complexity and implementation cost are lowered, and the system performance is improved.

Description

technical field [0001] The invention relates to the technical field of maximum power point tracking of photovoltaic systems, in particular to a multi-peak MPPT algorithm. Background technique [0002] In photovoltaic systems, in order to quickly track to the global maximum power point, improve tracking efficiency, reduce shocks, adapt to environmental changes, and avoid falling into local peaks, a series of multi-peak maximum power tracking algorithms have been proposed, mainly including the following categories of algorithms: [0003] (1) The multi-peak maximum power point tracking method of photovoltaic inverters. This method uses a large step to scan the local peak area. Once the local peak is scanned, use a small step to find the maximum power point of the local peak, and then continue Use a large-step scan. After the global scan, take the local maximum power point of the largest peak interval as the global maximum power point. In order to improve the scan efficiency, th...

AI Technical Summary

Problems solved by technology

[0009] (1) When the external environment changes in this method, if the maximum power point is not between 0.6 times Voc (open circuit voltage) and 0.9 times Voc, the phenomenon that GMPP cannot be traced will occur, resulting in a large amount of waste of electric energy; when the peak value of the system When it is very large, the algorithm needs to use a small step size to calculate the size of each local peak, which will affect the tracking speed of the algorithm;

[0010] (2) The multi-peak MPPT algorithm based on global scanning needs to find the maximum power point of each local peak, and its speed is relatively slow when there are many peaks; when the environment changes, the phenomenon of misjudgment of the maximum power point will occur

[0011] (3) The particle swarm optimization algorithm itself is relatively complex, and there will be some oscillations and unstable convergence near the maximum power

[0012] (4) The design of the fuzzy subset in the fuzzy logic algorithm requires artificial experts to design, has no evolution ability, and is not easy to expand to other photovoltaic systems

[0013] (5) The structure of the neural network algorithm is relatively complex, the hardware requirements are relatively high, and it also requires a long training time, so it is not easy to expand to other photovoltaic systems

[0014] In large-scale photovoltaic systems, its P-U (power-voltage) characteristic curve often presents multi-peak characteristics, causing the traditional single-peak MPPT (Maximum Power Tracking) algorithm to easily fall into local peaks and seriously reduce system power generation efficiency. Therefore, it is necessary to study more Peak MPPT Algorithm

After a lot of research on MPPT algorithms in photovoltaic systems, it is found that most multi-peak MPPT algorithms have problems such as insufficient tracking speed, oscillation of the maximum power point, misjudgment, inability to quickly respond to dynamic changes in the environment, complex algorithms, or high implementation costs.

Images