摘要
针对动态非线性、时变发酵过程,采用混合核支持向量机的智能模型建模方法,通过建立混合核支持向量机的状态预估模型,实现对发酵产物浓度的预估,解决了缺乏生物传感器的问题。在此基础上,再利用粒子群优化算法求取补料速率优化曲线,最终使得发酵终止时产物产量最高。实验结果表明,该方法取得了良好的效果。
In according to the features of non-liner and time varying for the ferment process, using a intelligent modeling method of mixture kernels SVM, by the establishment of MKSVM model, the concentration of fermentation product can be estimated, it also solve the problem of lacking bio-sensor. Based on this model, particle swarm optimization（PSO） is applied to seek the optimal curve of material makeup rate, eventually making the product of fermentation production to the highest. Experimental results show that the method achieved good results.
出处
《自动化与仪表》
北大核心
2009年第5期23-27,共5页
Automation & Instrumentation
基金
国家863计划项目(2006AA020301)
关键词
支持向量机
粒子群算法
发酵建模
补料
support vector machine（SVM）
particle swarm optimization（PSO）
modeling of fermentation
material making-up
