A Novel Wind Power Prediction Scheme by Coupling the BP Neural Network Model with the Fireworks Algorithm

Main Article Content

Yonggang Li
Yaotong Su
Lei Xia
Yongfu Li
Hong Xiang
Qinglong Liao


Wind power has unpredictable, intermittent traits due to meteorological conditions and environmental factors. Large-scale grid integration of wind energy will undoubtedly challenge system stability. This study developed a fireworks algorithm-backpropagation (FWA-BP) neural network model to forecast wind power using wind speed, direction, and power as model inputs. Optimization of the BP network weights and thresholds occurred through the fireworks algorithm. Compared to a standard BP network, the FWA-BP model yielded improved prediction accuracy seen through a lower mean squared error. This implies that the approach introduced in this paper significantly enhances global search capabilities, prediction accuracy, and speed. It contributes to enhancing the reliability of the power system, optimizing resource allocation, and improving wind power scheduling, with substantial potential and economic significance.

Article Details

Special Issue - Efficient Scalable Computing based on IoT and Cloud Computing