摘要
针对发酵过程中生物参数难以实时在线测量的问题,建立了用于生物参数状态预估的支持向量机软测量模型。考虑到该支持向量回归(SVR)模型的复杂性和冷化特征取决于其三个参数,εc,γ能否取到最优值,采用粒子群优化(PSO)算法实现对参数,εc,γ的同时寻优。在此基础上,以饲料用β-甘露聚糖酶为对象,建立了基于PSO-SVR的发酵过程产物浓度状态预估模型。发酵罐控制结果表明:该模型具有很好的学习精度和泛化能力,可实现对β-甘露聚糖酶产物浓度的实时在线预估。
In view of the hardship to get real-time and on-line biological parameters in fermentation process, a soft sensor model based on support vector machines is established for estimating the biological parameters. The complexity and generalization performance of the support vector regression （SVR） model depend on a good setting of the three parameters ε,c,γ. An algorithm called particle swarm optimization （PSO） is applied to optimize the three parameters synchronously. Based on the proposed method, a PSO-SVR model is developed to estimate the products concentration of beta- mannanase for feedstuff. The control results of fermenter show that the state estimation model based on PSO-SVR has good learning accuracy and generalization performance so as to obtain the real-time and on-line estimation for products concentration of beta-mannanase.
出处
《南京理工大学学报》
EI
CAS
CSCD
北大核心
2008年第4期517-521,共5页
Journal of Nanjing University of Science and Technology
基金
国家"863"计划(2003AA241160)
江苏省自然科学基金(BK2005012)
关键词
支持向量回归
状态预估
粒子群优化
发酵过程
Β-甘露聚糖酶
support vector regression
state estimation
particle swarm optimization
fermentation process
beta-mannanase
