Photovoltaic power plant equivalent modeling method based on improved fuzzy clustering algorithm

Abstract

The invention discloses a photovoltaic power plant equivalent modeling method based on an improved fuzzy clustering algorithm. The method comprises the steps of (1) determining a clustering indicatorof each photovoltaic power generation unit and performing standardization processing, (2) setting clustering parameters and initializing a membership degree matrix, (3) clustering photovoltaic power generation units by an improved fuzzy clustering algorithm, (4) calculating equivalent parameters according to a clustering result and combining the photovoltaic power generation units, and (5) verifying the correctness of an equivalent model by error analysis. According to the method, the photovoltaic power generation units can be scientifically and effectively grouped, and therefore, the accuracyof an equivalent model of a large-scale photovoltaic power station can be greatly improved.

Description

technical field [0001] The invention relates to the field of power system simulation analysis, in particular to an equivalent modeling method for a large-scale grid-connected photovoltaic power station. Background technique [0002] With the gradual depletion of global fossil energy, photovoltaic power generation has developed rapidly in countries all over the world due to its clean, non-polluting, renewable, and huge reserves. In recent years, photovoltaic grid-connected simulation analysis has become a research hotspot in the field of power systems. The traditional simulation method establishes detailed simulation models for all grid-connected photovoltaic power generation units to achieve accurate simulation of the transient steady-state performance of photovoltaic grid-connected. However, with the continuous expansion of the scale of photovoltaic power plants, for a large number of centralized photovoltaic grid-connected units, the traditional modeling method not only c...

AI Technical Summary

Problems solved by technology

In FCM, the sum of the membership degrees of the same sample belonging to all classes is 1, which makes it very sensitive to noise and outlier points; the iterative descent algorithm adopted by FCM is sensitive to the initialized cluster center or membership matrix, Can not guarantee to converge to the global optimal solution, may converge to local extremum or saddle point

Images