电镦机中加热电流的综合集成网络预报模型的算法研究

ALGORITHMIC RESEARCH ABOUT SYNTHETIC NEURAL NETWORK PREDICTION-MODEL OF THE HEATING CURRENT IN THE ELECTRIC UPSETTING

结合电镦机中加热电流的数学模型,详细分析了综合集成神经网络算法的实现过程,其仿真结果表明,它在“小样本”时,不仅能减少连接权值,而且能加快训练速度,提高泛化能力,实现了电镦机中的关键参数——加热电流的高精度预报。

Based on the mathematic model of heating current in the electric upsetting, the paper analyses the algorithmic implementation of synthetic neural network in detail, the results of the simulink could be shown that a synthetic method of neural network can reduce the number of network weight, improve the learning speed, generalization performance of the network in the lack of learning sample, as well as realize a high precision prediction of the heating current ,which is the key parameter in the electric upsetting.