基于支持向量机的发酵过程建模研究

发酵过程有众多关键性的变量难以在线检测，给过程优化策略的有效实施带来了障碍。最小二乘支持向量机（LS-SVM）是标准支持向量机（SVM）的一种扩展，LS-SVM算法精度高，速度快，适合于在线预估。将该算法用于青霉素发酵过程建模，用具有RBF核函数的LS-SVM建立菌体浓度、青霉素浓度的模型，并通过仿真实验与标准支持向量机进行比较。结果表明，最小二乘支持向量机是青霉素发酵过程建模与控制的一种有效的方法。

Modeling of Penicillin Ferment Process Based on Least Square Support Vector Machine

Many critical variables,are difficult to be obtained in fermentation process,which holds back the effective application of optimal procedure strategy.Least square support vector machine(LS-SVM) is a kind of extension of support vector machine(SVM).The algorithm of LS-SVM has high precision,fast computing.It is more suitable for predicting online.LS-SVM is introduced to model penicillin fermentation process.A model is built by LS-SVM with RBF kernel for mycelial and penicillin.The result shows that LS-SVM is an effective method for penicillin ferment process modeling and control.