：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 ，c， 能否取到最优值，采用粒子群优化(PSO)算法实现对参数 ，c， 的同时寻优。在

此基础上，以饲料用 ．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 ．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 bio—

logical parameters．The complexity and generalization performance of the support vector regression

(SVR)model depend on a good setting of the three parameters ，c， ．An algorithm called parti—

cle 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 feedstuf．The control results of fermenter show that the state estimation model based

on PSO·-SVR has good learn ing accuracy and generalization perform ance SO as to obtain the real·-time

and on—line estimation for products concentration of beta—mannanase．