A neural network photovoltaic power prediction model and method based on secondary dynamic adjustment

Abstract

The invention discloses a neural network photovoltaic power prediction model and method based on secondary dynamic adjustment. The prediction model is based on a resource allocation neural network, and uses historical data to offline training resource allocation network learning rules to obtain online prediction and background secondary dynamic adjustment. Initial neural network prediction model; put the initial prediction model into the actual photovoltaic power prediction, use real-time data as the model input, and record the samples with large deviations in the prediction results; when the prediction results with large errors appear again, they will be matched with the data in the buffer , if there is input data with similar features, increase the support of the current type of input samples; when the input samples with similar features meet the support threshold, enable secondary dynamic adjustment to adjust the structure of the prediction model to learn this type of samples. The invention solves the problem that the neural network prediction model solely relies on offline training samples, and at the same time enables the prediction model to have online learning ability and is more suitable for the characteristics of photovoltaic power.

Description

technical field [0001] The invention relates to the field of photovoltaic power generation power prediction, in particular to a neural network photovoltaic power prediction model and method based on secondary dynamic adjustment. Background technique [0002] Today, solar photovoltaic power generation has become an emerging industry that is widely concerned and focused on by countries around the world because of its cleanliness, safety, convenience, and high efficiency. It has also become an important part of new energy and renewable energy, and is developing rapidly. However, photovoltaic power generation is closely related to meteorological conditions, with uncertainty and volatility. This makes it very difficult to control the output of the photovoltaic system and maintain the supply and demand balance of the total power generation, which can easily affect the efficiency and reliability of the power system. In this case, PV power prediction is proposed as an economical an...

AI Technical Summary

Problems solved by technology

[0004] The present invention aims at the problem that the existing neural network photovoltaic power prediction model only relies on the characteristics of training samples, does not have online learning ability, and cannot adapt to the dynamic change characteristics of photovoltaic power, and proposes a neural network photovoltaic power prediction model based on secondary dynamic adjustment and methods

Images