应用加权LSSVM算法的PVC气提塔温度建模

Temperature Modeling of PVC Stripping Process Using Weighted Least Square Support Vector Machine

研究基于最小二乘支持向量机(LSSVM)建立了PVC汽提塔的预测模型。为了提高LSSVM的鲁棒性,过滤离群点,将加权最小二乘支持向量机(WLSSVM)应用到PVC汽提过程的温度建模中,对汽提塔温度进行建模和仿真实验。对比仿真实验结果表明:WLSSVM建模具有更高的建模精度和更优秀的性能。

The temperature control of the stripping process is particularly important.Considering that the temperature control has obvious non-linear characteristics,and support vector machine(SVM)has significant advantages in the nonlinear system control,this study establishes the prediction model of PVC stripper based on least square support vector machine(LSSVM).In addition,in order to improve the robustness of LSSVM and filter outliers,a weighted least squares support vector machine(WLSSVM)is applied to model the temperature of PVC stripping process.The above two methods are used to model and simulate the temperature of stripper.The results of simulation show that WLSSVM model has higher modeling accuracy and better performance.