1. Research on Marine Electric Load Forecast Based on PSO-Elman Neural Network
- Author
-
Shi Wei-feng, Bi Zong, Song Tiewei, and Xie Jialing
- Subjects
Local optimum ,Artificial neural network ,Electrical load ,Stochastic process ,Control theory ,Computer science ,Computer Science::Neural and Evolutionary Computation ,Marine energy ,Stability (learning theory) ,Electric power ,Function (mathematics) ,Physics::Atmospheric and Oceanic Physics - Abstract
The objective of this research is to improve the performance of marine electric power load forecasting based on PSO-Elman neural network. Considering the stochastic characteristics of marine power load variation, using the arbitrary approximation ability of Elman neural network, transform the prediction problem into a function fitting problem to establish a marine electric load forecasting model, and the Elman neural network learning method is optimized by particle swarm algorithm to avoid falling into local optimum. Simulation results obtained have shown that through the learning of a ship’s electrical load data, the load forecasting model based on PSO-Elman neural network can describe the changes in marine electrical load and improve the prediction accuracy and stability.
- Published
- 2021