Air Material Demand Forecast Based on Combined Neural Network

Accurate forecast of air material demand can not only improve the refinement degree of air material support, but also increase the predictability of air material support, and lay the foundation for completing various flight missions. This paper makes full use of the self-adaptive, self-organizing and self-learning ability of the artificial neural network, and puts forward a combined forecasting model based on LVQ neural network, Elman neural network and SOM neural network. Using entropy theory, the weight coefficients of each forecast method are determined. The example proves that the method has good effect.