[1]嵇小辅,张翔.基于FCM与集成高斯过程回归的赖氨酸发酵软测量[J].智能系统学报,2015,10(1):156-162.[doi:10.3969/j.issn.1673-4785.201310070]
JI Xiaofu,ZHANG Xiang.Soft measurement of lysine fermentation based on FCM and integrated Gaussian process regression[J].CAAI Transactions on Intelligent Systems,2015,10(1):156-162.[doi:10.3969/j.issn.1673-4785.201310070]
点击复制
JI Xiaofu,ZHANG Xiang.Soft measurement of lysine fermentation based on FCM and integrated Gaussian process regression[J].CAAI Transactions on Intelligent Systems,2015,10(1):156-162.[doi:10.3969/j.issn.1673-4785.201310070]
基于FCM与集成高斯过程回归的赖氨酸发酵软测量
《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷:
10
期数:
2015年第1期
页码:
156-162
栏目:
学术论文—机器学习
出版日期:
2015-03-25
- Title:
- Soft measurement of lysine fermentation based on FCM and integrated Gaussian process regression
- 关键词:
- 高斯过程回归(GPR); 模糊C均值聚类(FCM); Adaboost算法; L-赖氨酸; 软测量; 欧氏距离; 隶属度; 加权求和
- Keywords:
- Gaussian process regression(GPR); fuzzy C-mean clustering (FCM); Adaboost algorithm; L-lysine; soft measurement; Euclidean distance; membership; weighted sum
- 分类号:
- TP274
- 文献标志码:
- A
- 摘要:
- 为解决赖氨酸发酵过程中菌体浓度难以在线检测的难题,提出一种基于模糊C均值聚类(FCM)与集成高斯过程回归(GPR)的软测量建模方法。针对典型生物发酵过程可分为延滞期、指数生长期、稳定期、死亡期4个反应周期的特点,采用模糊C均值聚类算法对样本集进行聚类分析以形成若干子样本集;对每个子样本集分别采用高斯过程回归训练时,为提高GPR模型的泛化能力,利用Adaboost算法提升GPR模型,分别在各子集建立集成GPR软测量子模型;采用欧氏距离计算新样本点对应于每一子模型的隶属度;加权求和获得最终的软测量模型的预测输出。基于氨基酸类典型菌种L-赖氨酸反应过程菌体浓度参数预测的试验研究表明:与全局单一GPR模型、集成GPR模型和基于FCM与多GPR模型相比,所建立的基于FCM与集成GPR软测量模型拟合精度高,泛化能力强,较好地满足了赖氨酸发酵过程的控制要求。
- Abstract:
- In order to solve the problem that cell concentration is difficult to directly measure in the lysine fermentation process, a kind of soft measurement modeling method is proposed on the basis of fuzzy C-mean clustering (FCM) and integrated Gaussian process regression (GPR). The characteristics of typical biological fermentation process can be divided into 4 reaction cycles, including lag phase, exponential growth phase, stable phase, and dead phase. The cluster analysis is conducted for a sample set by applying fuzzy C-mean clustering algorithm, so as to form several sub-sample sets. In order to improve the generalization performance of the GPR, each group is trained through Gaussian Process Regression based on Adaboost and the corresponding integrated sub-models are established. The memberships between each new sample and each group are set as the weights through Euclidean distance and the predicted result is obtained by weighted sum by using typical bacterium of amino acid—L-lysine fermentation as an example. The simulation results showed that compared with the global single GPR model, integrated GPR model and the model based on FCM and multiple GPR, the soft measurement model based on integrated GPR and FCM has high fitting precision. It also had strong generalization ability, which meets the control requirements of lysine fermentation process.
- 参考文献/References:
-
[1] BROSILLOW C B. Inferential control of process[J]. Automatica,1978,24(3):485-509.
[2] RASMUSSEN C E, WILLIAMS C K. Gaussian processes for machine learning[M]. Boston: MIT Press,2006: 7- 31.
[3] KOCIJAN J. Control algorithms based on Gaussian process models:a state-of-the-art survey[C]//Proc of the Special International Conference on Complex Systems:Synergy of Control,Communications and Computing. Ohrid, 2011:69-80.
[4] PETELIN D, KOCIJAN J, GRANCHAROVA A. Online Gaussian process model for the prediction of the ozone concentration in the air[J]. Comptes Rendus del Academie Bulgare des Sciences,2011,64(1):117-124.
[5] HE Z K, LIU G B,ZHAO X J, et al. Temperature model for FOG zero-bias using Gaussian process regression[J].Advances in Intelligent Systems and Computing,2012,180:37-45.
[6] 孙斌,姚海涛,刘婷.基于高斯过程回归的短期风速预测[J]. 中国电机工程学报, 2012,32(29):104-108.SUN Bin,YAO Haitao, LIU Ting. Short-term wind speed forecasting based on gaussian process regression[J]. Proceedings of the CSEE, 2012,32(29):104-108.
[7] BEZDEK J C. Pattern recognition with fuzzy objective function algorithms[M]. New York:Plenum Press,1981:15-28.
[8] 曲福恒,胡雅婷,马驷良,等.基于核的模糊C均值聚类算法的收敛性定理[J].吉林大学学报:理学版,2011,49(6): 1079-1086.QU Fuheng,HU Yating,MA Siliang, et al.A convergence theorem of kernel based fuzzy C-means clustering algorithm[J].Journal of Jilin University: Science Edition,2011,49(6): 1079-1086.
[9] MOHAMMAD T K,MOHAMMAD H B.A powerful hybrid clustering method based on modified stem cells and fuzzy C-means algorithms[J]. Engineering Applications of Artificial Intelligence,2013,26,1493-1502.
[10] XIAO Fengyin,LI P K,YIH T C. A fuzzy C-means based hybrid evolutionary approach to the clusering of supply chain[J]. Computers & Industrial Engineering,2013,66(4): 768-780.
[11] BENAICHOUCHE A N,OULHADJ H,SIARRY P. Improved spatial fuzzy C-means clustering for image segmentation using PSO initialization,Mahalanobis distance and post-segmentation correction[J]. Digital Signal Processing,2013,23(5): 1390-1400.
[12] PENG X J,SETLUR S,GOVINDARAJU V,et al.Using a boosted tree classifier for text segmentation in hand-annotated documents[J]. Pattern Recognition Letters, 2012,33(7): 943-950.
[13] 曹莹,苗启广,刘家辰,等. Adaboost算法研究进展与展望[J]. 自动化学报,2013,39(6):745-758. CAO Ying,MIAO Qiguang,LIU Jiachen, et al.Advance and Prospects of AdaBoost Algorithm[J]. Acta Automatica Sinica,2013,39(6):745-758.
[14] CHENG Wenchang, DING M J. A self-constructing cascade classifier with AdaBoost and SVM for pedestrian detection[J]. Engineering Applications of Artificial Intelligence,2013,26(3): 1016-1028.
[15] VALIANT L G. A theory of the learnable[J]. Communications of the ACM,1984,27(11):1134-1142.
[16] 仲蔚,俞金寿.基于模糊C均值聚类的多模型软测量[J]. 华东理工大学学报, 2000,26(1): 83-87.ZHONG Wei,YU Jinshou.Study on soft sensing modeling via FCM based multiple models[J]. Journal of East China University of Science and Technology,2000,26(1): 83-87.
[17] 黄永红,孙玉坤,王博,等. 赖氨酸发酵过程关键参数的模糊神经网络逆软测量研究[J]. 仪器仪表学报,2010,31(4): 862-867.HUANG Yonghong,SUN Yukun,WANG Bo, et al.Research of soft sensor based on fuzzy neural network inverse system for lysine fermentation process[J]. Chinese Journal of Scientific Instrument,2010,31(4): 862-867.
[18] 王博,孙玉坤,嵇小辅. 基于PSO-SVM逆的赖氨酸发酵过程软测量[J].化工学报,2012,63(9):3000-3007.WANG Bo,SUN Yukun,Ji Xiaofu, et al. Soft-sensor modeling for lysine fermentation processes based on PSO-SVM inversion[J]. CIESC Journal,2012,63(9):3000-3007.
[19] 孙玉坤,王博,黄永红.基于聚类动态LS-SVM的L-赖氨酸发酵过程软测量方法[J].仪器仪表学报,2010,31(2): 404-409.SUN Yukun,WANG Bo,HUANG Yonghong,et al.Soft-sensing method for L-lysine fermentation process based on FDLS-SVM[J]. Chinese Journal of Scientific Instrument,2010,31(2): 404-409.
[20] 黄丽,孙玉坤,嵇小辅,等. 基于tPSO-BPNN的赖氨酸发酵软测量[J]. 仪器仪表学报,2010,31(10):2317-2321.HUANG Li,SUN Yukun,JI Xiaofu, et al. Soft sensor of lysine fermentation based on tPSO-BPNN[J]. Chinese Journal of Scientific Instrument,2010,31(10):2317-2321.
备注/Memo
收稿日期:2013-10-27;改回日期:。
基金项目:国家863计划资助项目(2011AA09070301);江苏高校建设优势学科工程资助项目(苏政办发[2011]6号);江苏省科技支撑计划资助项目(BE2010354);江苏省自然科学基金资助项目(BK2011465).
作者简介:嵇小辅,男,1979年生,副教授,博士,主要研究方向为生物反应过程软测量与优化控制、鲁棒控制;张翔,男,1988年生,硕士研究生,主要研究方向为生物反应过程软测量与优化控制。
通讯作者:张翔.E-mail:zhangxiang_mail@126.com.
更新日期/Last Update:
2015-06-16