The best way to conference proceedings by Francis Academic Press

Web of Proceedings - Francis Academic Press
Web of Proceedings - Francis Academic Press

Forecasting the Electricity Production from Renewable Energy in Shanghai

Download as PDF

DOI: 10.25236/scmc.2019.023

Author(s)

Yingjie Tian, Naiwang Guo

Corresponding Author

Yingjie Tian

Abstract

The development of new and renewable energy in China is still in its infancy. Comprehensive consideration can confirm that there is great potential for future development, but there are also many problems restricting development. From the point of view of the development of new technologies, this paper applies cluster analysis to the field of photovoltaics, and establishes a prediction model of photovoltaic power generation based on neural network. Aiming at the three weather types of Shanghai sunny, cloudy and rainy days, the abnormal samples in historical data are screened by cluster analysis, and the back propagation (BP) neural network prediction model is established by using the selected samples as training data. By comparing the results of the forecasting models before and after screening, it can be seen that the forecasting model based on the data filtered by clustering analysis has higher accuracy. Therefore, the combination of clustering analysis and BP neural network is an effective method to improve the forecasting accuracy of photovoltaic power generation.

Keywords

Renewable Resources; Cluster Analysis; Data Screening; Neural Network