论文部分内容阅读
传统的支持向量回归算法因基于批量训练方法而无法适应浸出过程在线建模实时性的要求.在分析研究一种基于矢量基学习的支持向量回归算法的基础上,提出了基于矢量基学习的浸出过程在线建模方法.利用贝叶斯证据框架优化模型参数,分析新样本矢量与矢量空间的夹角,从而推导出该样本是否为基矢量.将该方法应用于浸出过程浸出率的预测,实验结果表明,该方法不但能很好地跟踪浸出率的变化趋势,而且显著地缩短了运算时间.
The traditional support vector regression algorithm can not meet the real-time requirement of on-line modeling of the leaching process because of the batch training method.Based on the analysis of a support vector regression algorithm based on vector-based learning, this paper proposes a vector-based learning leaching Process online modeling method.By using the Bayesian framework to optimize the model parameters, the angle between the new sample vector and the vector space is analyzed to deduce whether the sample is the base vector or not.The method is applied to predict the leaching rate in the leaching process, and the experiment The results show that this method can not only keep track of the trend of leaching rate, but also shorten the calculation time significantly.