基于多输出支持向量机的聚乙烯相对分子质量分布建模与优化(英文)

Modeling and optimization for molecular weight distribution of polyethylene based on multi-output support vector machine

相对分子质量分布(MWD)是聚合物产品重要的质量指标,目前实时检测相对分子质量分布仍缺乏有效的方法.在聚合物研究中,高精度的相对分子质量分布实时检测方法是当前的研究热点.不同于其他文献中描述的方法,通过结合聚合反应机理和过程信息建立相对分子质量分布的混合模型来解决预测精度和实时性问题.首先利用催化剂各活性中心的分布函数加权叠加拟合相对分子质量分布,分布函数参数和工艺条件之间的关系可通过多输出支持向量机回归(MSVR)算法来描述;其次应用无约束非线性优化方法优化上述方法建立的混合模型的工艺条件;最后将建立的混合模型应用于乙烯聚合过程,验证了上述方法的可行性.

The molecular weight distribution （MWD）is an important quality index for the polymer materials,but detecting the MWD in real-time is still difficult up to now.The real-time detection method for the MWD with good accuracy is hotspot in the polymer production.Different from the methods other literatures have described,the reaction mechanism and the industrial information are merged to build a hybrid model for the MWD to solve the prediction accuracy and real-time problems.The weighted superposition of the distribution function on each active center of catalyst is used to fit the MWD. The multi-output support vector machine regression （MSVR ） algorithm is applied to the description of the relationship between process conditions and parameters of the distribution function.Then,the unconstrained nonlinear optimization method is used to optimize the process conditions based on the hybrid model.Finally,the application of the above-mentioned approach in the ethylene polymerization process verifies its feasibility.