The Research And Application Of Hybrid Model Based On Elman Neural Network And Optimization Algorithm

Artificial neural networks are mathematical and physical methods as well as from the perspective of the human brain information processing biological neural network model is a simplified abstraction built up in recent years in theoretical research, applied research and implementation techniques and achieved remarkable achievement. Especially neural networks and genetic algorithms, particle swarm optimization, ant colony algorithm, fuzzy theory, intelligent information processing methods, such as the combination of chaos theory, which form a "hybrid neural network model," is currently a major research focus in the field of artificial intelligence. Artificial intelligence and neural network theory and algorithms combined with each other, has been widely used in pattern recognition, signal processing, control and optimization of practical engineering, communications transmission, especially in the forecasting model.This article describes some of the basics of neural networks and artificial intelligence optimization algorithms, and details a dynamic feedback mechanism of the artificial neural network model-Elman neural network and two hybrid neural network model-genetic neural network, PSO neural networks and their mathematical models and learning algorithms.Taking into account the wind speed forecasting plays an important role in the wind energy, whether wind or wind energy resource assessment and network planning and power system operators have a critical impact, so the wind speed forecasting of wind farms has important application value. Because many factors affect the wind speed change, and change in wind speed with a strong non-linear characteristic. Neural network has strong nonlinear mapping function, we use the Elman neural network hybrid wind speed of a wind farm in Chengde make short-term forecasting, simulation results and error analysis results show that hybrid forecasting model based on Elman neural network can effectively improve the prediction accuracy of the wind speed, prove the feasibility of this model.