支持向量機(jī)（SVM）是一種基于結(jié)構(gòu)風(fēng)險(xiǎn)最小化原理(SRM)的學(xué)習(xí)算法，也是一種具有很好的泛化性能的回歸方法。針對(duì)青霉素發(fā)酵過(guò)程中的菌體濃度進(jìn)行軟測(cè)量建模，提出了一種新的基于距離的模糊支持向量機(jī)，并用序列最小優(yōu)化算法(SMO)求解優(yōu)化問(wèn)題。仿真實(shí)例說(shuō)明能夠?qū)η嗝顾匕l(fā)酵過(guò)程中不可在線(xiàn)測(cè)量的變量進(jìn)行軟測(cè)量，達(dá)到了較高的測(cè)量精度。
關(guān)鍵詞: 支持向量機(jī)；模糊；菌體濃度；距離；序列最小優(yōu)化；

Application of Fuzzy Support Machine in Penicillin Ferment Zhenya Yu, Wenxia Wang, Feng Pan （School of Communication and Control Engineering, Southern Yangtze University, Wuxi 214122） Abstract: Support vector machine (SVM) is a learning algorithm based on structural risk minimization (SRM), it is a good regression way, which has good generalization ability also. A fuzzy support vector machine based on distance which used to model the mycelial concentration by soft sensor is proposed in this article. The optimization question is resolved by sequential minimal optimization (SMO). The simulation example shows that this FSVM could measure the parameters, which could not be measured online during the course of penicillin fermentation, with a high precision.
Keyword: support vector machine; fuzzy; mycelial concentration; distance; sequential minimal
optimization