【摘要】 在对样本数据进行预处理的基础上,建立一个具有自适应残差补偿的改进BP神经网络动态预报模型,并对神经网络的学习参数进行自适应调整。将该模型应用于铜锍吹炼过程所需的氧气量进行预报。仿真结果表明,预报最大相对误差为3.97%,最小相对误差可以达到0.11%。该模型已应用于实际生产,具有精确度高、实用的优点。更多还原

【Abstract】 After a data set is preprocessed, a dynamic prediction model based on an improved BP neural network and error compensation of linear regression has been proposed, and the learning parameters are adjusted adaptively. This model is applied to predict oxygen volume of copper smelt converting. The simulation result shows that the minimum error of the model is 0.11%, and the maximal error of the model is 3.97%. The model is proved to be precise and practical.更多还原