【摘要】 由于传统的轧制压力计算模型结构简单，即使采用自适应技术，也难以适应不断提高的尺寸精度的要求。为了提高精轧机组轧制压力预设定精度，采用人工神经网络和传统数学模型相结合的综合神经网络方法。经离线仿真证明，该方法的预报精度优于传统方法，预报结果的相对误差基本限制在±５％以内。更多还原

【Abstract】 As the structure of classical mathematical model is simple, even with selfadapting technology, it is difficult to accommodate the increasing dimensional accuracy. Motivated by this fact, it is discussed that the neural networks in combination with mathematical models are used to improve the precision of prediction of rolling load. Offline simulation shows that aforementioned method is much better than the classical one. The relative error is between ±5 %.更多还原