基于自适应模糊系统加动量项的BP神经网络的铜转炉模型

he copper coverter model based on the production data ,the BP neural network and the adaptive fuzzy system,the model can forcast the content of magnetic oxide and copper in the slag.This model makes up for the shortcomings of the traditional BP neural network that easy to be immersed in the minimum point and the training without convergence.This model combines fuzzy inference function in order to Strengthening BP algorithm.So it improved the rate of BP algorithm and the accuracy of prediction is strengthened.By comparing the predicted results with the actual production results. It shows that the range of error is reasonable.Through the fine prediction function, the model can optimize the charge composition of the production of copper smelting converter.It not only savesthe sources of energybut also ruduce the content of magnetic oxide and copper in the slag.