A high-precision dynamic modeling method for photovoltaic power plant cluster

Abstract

The invention discloses a high-precision dynamic modeling method of a photovoltaic power station cluster. On the basis of analyzing the limitation of the existing clustering equivalent modeling method, an error correction method is proposed, and a long-short-time memory (LSTM) network is introduced to constitute an error correction system to model the error of the existing clustering equivalent model. The invention synthesizes an error correction system and an equivalent method based on clustering, and finally proposes a depth learning-clustering(DL clustering) hybrid modeling framework. The hybrid modeling framework provided by the invention greatly improves the accuracy of the equivalent model while retaining the advantages of low order of the clustering equivalent model and less simulation time consumption. The framework of the invention can be extended and applied to the modeling of other high-permeability renewable energy distribution network and the like.

Description

technical field [0001] The invention belongs to the technical field of renewable energy grid-connected modeling and simulation, and in particular relates to a high-precision dynamic modeling method for photovoltaic power station clusters. Background technique [0002] In recent years, with the increasing dependence of human society on energy and the depletion of traditional fossil energy, renewable energy such as photovoltaics, which has the advantages of low-carbon environmental protection and abundant resources, has become an important means of energy development in many countries. In 2050, the proportion of photovoltaic power generation in the total global electricity consumption will reach 16%. Thanks to the gradual maturity of photovoltaic technology and the encouragement of government policies, small and medium-sized photovoltaic power plants with a two-stage structure appear in large numbers and concentratedly in application environments such as the roofs of industria...

AI Technical Summary

Problems solved by technology

[0005] However, in practice, the fitting performance of the cluster equivalent model of this photovoltaic cluster to the dynamic characteristics will decrease as the dynamic parameters of different photovoltaic power plants in the cluster tend to be evenly distributed

This can easily lead to the fact that the accuracy of the clustering equivalent model cannot meet the needs of practical applications in more general cases, and this key defect is caused by the inherent properties of the clustering algorithm, which is difficult to achieve by improving the algorithm itself, so that It largely limits the application of cluster equivalent modeling methods

Images