中文摘要：

针对面向多输出系统支持向量机回归算法训练时间较长的问题,提出一种面向多输出系统的启发式支持向量机回归算法。与多输出的支持向量机回归建模相比,该方法建立的模型结构较为简单,模型训练速度更快。将此方法和直接支持向量机回归算法分别应用到甲基丙烯酸甲酯的间歇聚合反应过程中,仿真结果表明了该方法的有效性。

英文摘要：

Since training of the support vector regression algorithm for multi-output systems takes a long time,we have proposed a heuristic support vector regression algorithm for multi-output systems.Compared to regressive modeling of the multi-output support vector,the model created by our method is simpler with faster model training.The effectiveness of the new method was validated by simulation results when applied to the batch polymerization reaction process for methyl methacrylate.The new model was compared with the conventional support vector regression algorithm.