论文部分内容阅读
采用ＢＰ神经网络方法预报热连轧精轧机组轧制力·通过训练数据预处理、利用遗传算法优化网络结构和参数、按钢种划分训练样本等方法，提高了网络的预报精度，优于传统的数学模型方法·ＢＰ神经网络与数学模型相结合的综合神经网络方法，进一步提高了轧制力的预报精度·预测结果与实测数据比较表明，相对误差基本在±７％以内，实现了精轧机组轧制力的高精度预报
The BP neural network method is used to predict the rolling force of the hot strip mill. The forecasting accuracy of the network is improved by training data preprocessing, using genetic algorithms to optimize the network structure and parameters and classifying training samples according to the steel grades, Of the mathematical model method · BP neural network and mathematical model of the integrated neural network method to further improve the prediction accuracy of rolling force prediction results compared with the measured data show that the relative error of basic within ± 7% High precision forecast of unit rolling force