基于GMDH神经网络的轮胎硫化温度预测

介绍硫化过程的测温难点及GMDH神经网络的算法和构造。采用GMDH神经网络对硫化过程中轮胎内部温度进行预测，并与BP神经网络预测进行对比。研究结果表明，GMDH神经网络的温度预测精度较高，对非线性数据有很好的预测效果。采用该方法可为轮胎硫化控制提供有力支持，并实现轮胎生产无损测量。;The difficulty of temperature measurement for vulcanization process，and the arithmetic and construction of GMDH neural network were introduced.The temperature inside tire was predicted by using GMDH neural network，and compared with the prediction of BP neural network.The results showed that GMDH neural network had higher accuracy of temperature prediction，and was very efficient for nonlinear data.The GMDH neural network method could provide strong support to vulcanization process control and building-up of nondestructive temperature measurement method for tire production