Method for photovoltaic power station group region power prediction on basis of improved RBF neural network

Abstract

The invention discloses a method for photovoltaic power station group region power prediction on the basis of an improved RBF neural network. The method comprises the steps of dividing a whole region into a plurality of sub-regions, performing data quality control, and removing data points remarkably not conforming to practice; calculating relative coefficients between photovoltaic power station measured power in every sub-region and sub-region total measured power, and selecting a plurality of standard photovoltaic power stations for every sub-region; utilizing a method of combining physics and statics to achieve short-term power prediction of the standard photovoltaic power stations; establishing a particle swarm optimization RBF neural network model based on a genetic operator, and predicting short-term power of every sub-region; and accumulating power prediction results of every sub-region to obtain prediction total power of a region photovoltaic power station group. The method aims to help a power scheduling department to formulate next day generation schedules according to the region power prediction results, real-time scheduling adjustment is optimized, spinning reserve capacity of a power system is reduced, system running costs are reduced, and the system photovoltaic acceptance capability is further improved.

Description

technical field [0001] The invention belongs to the field of photovoltaic power prediction, and in particular relates to a method for predicting regional power of photovoltaic power station groups based on an improved RBF neural network. Background technique [0002] Solar energy is an inexhaustible renewable energy with the characteristics of cleanness, safety, wide distribution and rich reserves. Photovoltaic power generation is a form of power generation that uses solar cells to directly convert solar radiation energy into electrical energy based on the principle of the photovoltaic effect. Under the pressure of today's global energy shortage and environmental degradation, photovoltaic power generation is gradually favored by various countries and has developed rapidly. However, due to the intermittent, volatile, and cyclical characteristics of photovoltaic power generation, large-scale photovoltaic power generation grid integration will inevitably have a huge impact on ...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a method for forecasting regional power of photovoltaic power station groups based on improved RBF neural network to solve the problem of photovoltaic power generation power prediction in the entire region, especially when the power of a single photovoltaic power station cannot be predicted. prediction problem

Images