同时进行发酵和分离的CO_2气提、活性炭吸附乙醇发酵动力学研究——(二)游离细胞连续发酵

本文研究了游离细胞连续发酵时应用 CO_2气提、活性炭吸附的乙醇发酵和分离动力学,考察了稀释率和进口葡萄糖浓度对发酵速率的影响,提出了包括 CO_2汽提乙醇速率的发酵动力学模型,研究了连续发酵时的稳态和动态过程,结果表明,CO_2气提有助于减少乙醇对酵母的抑制作用,发酵动力学模型可以较好地关联稳态时的发酵动力学数据,从中获得的模型参数可以用于预测动态过程中底物、微生物及产物的变化规律。对于所研究的过程,存在着最佳稀释率和进口糖浓度以获得最大发酵速率。     

基于流变特性的动力学模型研究及在柠檬酸发酵过程的应用

研究了基于流变特性的动力学模型,并以实验统计数据为基础,通过对模型进行分析,推断了模型参数,同时将模型应用于柠檬酸发酵过程中。结果表明模型的拟合效果良好,从而表明所建立的模型基本符合柠檬酸发酵过程,这为显示流变特性对发酵代谢过程的影响提供了理论依据,也为实现柠檬酸发酵过程最优化提供必要条件。     

肌苷发酵和分离耦合过程的动力学

许多发酵过程都受到高浓度产物的抑制,肌苷发酵亦存在明显的产物反馈抑制作用。为了降低产物抑制效应,采取发酵和离子交换分离产物、流加补料结合起来的耦合发酵方法。本文结合肌苷合成代谢途径,建立了肌苷耦合发酵动力学模型,用于指导工业生产。     

基于分阶段的LSSVM发酵过程建模

发酵过程建模是研究微生物发酵的重要课题,基于模型可实现被测参量的软测量、系统的优化控制。鉴于引入混合核函数的最小二乘支持向量机在过程建模中具有优良表现,采用基于混合核函数的最小二乘支持向量机建模。但由于发酵过程周期较长,最小二乘支持向量机的全局模型预测精度难以保证,算法复杂度很高,因此提出一种分阶段建模方法。首先,选择表征阶段特性的辅助变量,利用模糊C均值聚类算法对样本数据聚类,将发酵过程分成不同的阶段,然后为各个阶段分别建立最优混合核最小二乘支持向量机局部模型,最后将局部模型合成构成过程的完整模型。将此方法应用于青霉素发酵过程和重组大肠杆菌发酵过程中,验证了该方法的有效性。     

诺西肽发酵过程中的分阶段软测量建模

诺西肽发酵过程中关键生化参数难以在线测量,给控制与优化带来困难.针对这一问题,利用软测量技术来实现关键生化参数的在线估计,并提出了一种分阶段软测量建模方法.首先以分阶段的诺西肽发酵过程非结构模型为基础,根据隐函数存在定理进行辅助变量的合理选择;然后利用模糊c均值聚类算法将建模数据按其所属阶段的不同进行分类,并利用神经网...     

基于PLS-LSSVM的谷氨酸发酵产物浓度预测建模

针对谷氨酸发酵过程关键生化参数难以在线检测给发酵优化控制带来困难问题,基于谷氨酸5 L发酵罐发酵过程,建立基于偏最小二乘（PLS）和最小二乘向量机（LSSVM）相结合的谷氨酸浓度预测模型;利用PLS对输入变量进行特征提取降低维数和消除相关性,以简化模型和提高模型精度。为确定谷氨酸发酵最佳预测模型,简化后的预测模型与发酵动力学模型进行比较;实验结果表明,简化后的耦合模拟退火（coupled simulated annealing,CSA）对参数进行优化的LSSVM模型具有最好预测性能,相对PLS预测模型和发酵动力学模型具有明显优势,均方根误差分别为1.597、8.49和2.934,可以为谷氨酸发酵过程操作及时调整及优化控制提供有效指导。     

Mathematical modeling of fed-batch butanol fermentation with simultaneous pervaporation

A mathematical model was developed to describe a fed-batch acetone-butanol-ethanol (ABE) fermentation with simultaneous pervaporation. The model predicted satisfactorily batch or fed-batch fermentation with or without pervaporalion by introducing a parameter reflecting cell activity loss during fed-batch fermentation with pervaporation. The model also predicted the effect of membrane area, membrane thickness, and sweep air flow rate on glucose consumption rate and residual butanol concentration in the fermentation broth. Glucose consumption rate increased by 30% by either doubling the membrane area or decreasing membrane thickness by half.     

放射型根瘤菌分批发酵生产辅酶Q10的代谢特性和发酵动力学

以放射型根瘤菌WSH2601作为辅酶Q10分批发酵的试验菌,对其代谢特性进行了初步研究。对分批发酵过程中细胞生长、产物积累、糖消耗、中间产物有机酸及pH的变化规律进行了描述：由Logistic模型方程分别建立了放射型根瘤菌辅酶Q10发酵过程细胞生长、产物合成及基质消耗随时间变化的数学模型。模型模拟计算结果与实验值能较好地吻合。动力学研究结果表明该模型能较好地反映放射型根瘤菌的细胞生长、底物消耗和产物合成过程及其动力学机制。辅酶Q10分批发酵中细胞生长与产物合成属于半偶联型。     

黄霉素工业化发酵过程中灰色预测模型的应用

将灰色预测模型应用于黄霉素工业化发酵过程中。对采用Streptomyces bambergiensis 在60 m3发酵罐中批式补料发酵生产黄霉素的过程分别建立GM（1,8）、GM（1,1）和灰色Verhulst模型。3种模型的预测模拟结果表明都可以模拟黄霉素工业化发酵过程,其平均相对误差分别为15.322%、6.093%和1.776%,残差平方和分别为4.177 g2/L2、0.680 g2/L2和0.053 g2/L2。通过比较3种模型预测模拟结果证明灰色Verhulst模型更适合黄霉素工业化发酵过程的预测模拟。     

基于pH控制的丙酮丁醇间歇发酵过程动力学模型

基于微生物发酵动力学模型和丙酮丁醇发酵工艺的特点,由5组控制pH下的实验数据建立该发酵过程的动力学模型,采用改进的自适应遗传算法对模型的12个参数进行优化,并采用一组控制pH数据对模型进行了外推验证,结果表明,该模型能够较好地描述pH的丙酮丁醇间歇发酵过程.     

与机理杂交的支持向量机为发酵过程建模

针对生物发酵过程机理复杂、高度非线性的特点,采用基于结构风险最小的支持向量机为发酵过程建模,其算法规范,建模复杂度低于神经网络方法,所建模型的预测效果更好.还将生化过程的动力学机理与支持向量机相结合,采用串联和串并联结构,提出与机理杂交的支持向量机建模方法,并为间歇式酒精发酵过程中酵母菌体浓度变化建立了预测模型.原理分析与试验结果表明与机理杂交的支持向量机建模方法,相比于单一近似的动力学模型、单一的支持向量机模型,以及机理杂交的神经网络模型,它的预测精度高,泛化能力强,性能更为优越.     

基于神经网络的发酵时间和最优发酵温度模型

以提高间歇式微生物发酵的产品得率为目标 ,利用BP神经网络和本文提出的傅立叶神经网络 ,提出发酵过程的发酵时间模型和最优发酵温度模型 ;在此基础上 ,提出了针对不同生产批次采用不同的最优发酵温度的新方法 ,此方法使不同生产批次的发酵过程都可以在适合其自身的最优的发酵温度下进行发酵 ,从而提高发酵生产的得率。实践应用表明 ,采用此种生产方案 ,产品平均得率提高 5 %。     

农业秸秆湿干两级厌氧发酵制沼气技术

干式发酵过程由于需水量少而成为目前国内外沼气行业的一个重要发展趋势,但是干式发酵存在启动慢和传质不均匀的局限性。采用湿式发酵作为干式发酵的接种启动阶段,将湿式和干式联合,从而力求解决干式发酵的局限性,发展出湿干结合的两级厌氧发酵产沼气工艺,以提高干式发酵技术的效率。采用秸秆为底物进行两级厌氧发酵,考察了第一级湿式发酵时间对后续干式发酵的影响。设置3、5、10、15和25 d五个不同湿式发酵周期进行实验,结果表明,湿式发酵周期为10、15和25 d实验组的累计两级甲烷产气能力相当,分别为218.63、219.44和218.85 ml·（g VS）^-1,3 d[208.17 ml·（g VS）^-1]和5 d[185.83 ml·（g VS）^-1]实验组产气量均较低。从产气量、降解程度和实际工艺等角度综合分析得出,湿式发酵周期为10 d之后转入第二级干法发酵效果最佳。动力学拟合分析表明,RC（reaction curve）模型较GM（Modified Gompertz）和LM（Modified Logistic）模型更适用于本实验研究的厌氧发酵过程的数据模拟。     

EXTRACTIVE FERMENTATION OF PHENYLALANINE USING UNCHARGED POLYMERIC BEADS

Extractive fermentation was employed for the production of phenylalanine using an agitated and aerated fermenter. Uncharged polystyrene beads (Amberlite XAD-16) were packed in an external column and used to extract the phenylalanine in an intermittent fashion during the fermentation. Even though no cell separator was used and the fermentation broth was fed directly to the column, the overall system performed well without any clogging of the column. It was found that 43% of the phenylalanine could be extracted during the fermentation, and it was thus proved that extractive fermentation is an effective way to prevent the crystallization of phenylalanine. In addition, fermentation performance as also improved due to the relief of self-inhibition by phenylalanine. When 1.2 L of the beads was used and sufficient hydrophobic nutrients were present, the concentration and productivity increased 18% and 55% greater than in the control fermentation, respectively. These results show that extractive fermentation is a very attractive method to enhance production as well as to avoid crystallization. Numerical simulation modeling was also performed in this study for extractive fermentation. The results from the modeling corroborated the experimental data, thus confirming the reliability of the model.     

A novel kinetic model to describe 1,3-propanediol production fermentation by Clostridium butyricum

Clostridium butyricum is one of the best 1,3-propanediol producers due to the nonpathogenic, less byproducts, and energy-efficient fermentation process. In fermentation process, the relationship among substrate, product, and byproducts is intricate and hard to be analyzed. The present study is aimed at establishing a novel kinetic model not only based on biomass, substrate, and 1,3-propanediol, but also considering the byproduct concentration to describe 1,3-propanediol fermentation process by C. butyricum. The simulative result of the model fit well with that in the batch fermentation process. Furthermore, the model was also used to predict the result of fed-batch fermentation process after some modifications. The predicted result of model fit well with the data in experiment when glycerol was controlled at around 10 g/L. Thus, this novel kinetic model could serve as a tool for further optimization of the fermentation process, and could be improved for some other similar processes.     

Extractive fermentation systems for organic acids production

Basic characteristics of on-line extractive fermentation of organic acids were examined using a general model for the integrated process in order to illustrate the effects of various process parameters and operational modes on system performance. The strong pH dependence of both the fermentation kinetics and the extraction efficiency has been outlined and taken into account by inclusion of a model equation which predicts the pH profile. Simulation results using data from butyric acid fermentations show that our complete model system is adequate to evaluate different operational modes of the process, including simple batch fermentation, batch or fed-batch fermentation with extractive recycle, and continuous fermentation with extractive recycle. In the case where a second undesired acidic byproduct is also produced, our model predictions suggest that on-line extractive fermentation using a suitable solvent results in an effective gross separation of the two acids.     

EXTRACTIVE FERMENTATION OF PHENYLALANINE USING UNCHARGED POLYMERIC BEADS

Extractive fermentation was employed for the production of phenylalanine using an agitated and aerated fermenter. Uncharged polystyrene beads (Amberlite XAD-16) were packed in an external column and used to extract the phenylalanine in an intermittent fashion during the fermentation. Even though no cell separator was used and the fermentation broth was fed directly to the column, the overall system performed well without any clogging of the column. It was found that 43% of the phenylalanine could be extracted during the fermentation, and it was thus proved that extractive fermentation is an effective way to prevent the crystallization of phenylalanine. In addition, fermentation performance as also improved due to the relief of self-inhibition by phenylalanine. When 1.2 L of the beads was used and sufficient hydrophobic nutrients were present, the concentration and productivity increased 18% and 55% greater than in the control fermentation, respectively. These results show that extractive fermentation is a very attractive method to enhance production as well as to avoid crystallization. Numerical simulation modeling was also performed in this study for extractive fermentation. The results from the modeling corroborated the experimental data, thus confirming the reliability of the model.     

粘质沙雷氏菌产灵菌红素的碳源分析、菌种诱变及动力学

使用流式细胞仪研究了不同碳源对粘质沙雷氏菌ZSG新陈代谢的影响,发现不同碳源导致ZSG的DNA含量、细胞内部颗粒密度、表面粗糙度和细胞大小在发酵过程中呈现有差异的变化。对ZSG进行复合诱变筛得一株稳定突变菌株ZSG7,在250 ml摇瓶和5 L发酵罐中进行发酵,其灵菌红素（PG）产量比出发菌株分别提高了62.5%和269%。对ZSG7进行发酵培养基优化后PG产量比优化前提高了100%。对ZSG7发酵进行溶氧分段控制模式调控后PG产量比DO调控前提高了30.9%。对ZSG7发酵进行恒定pH调控后比pH调控前提高了35.9%。对ZSG7发酵进行补料组分优化后比补料前提高了47.6%。基于Logistic方程和Luedeking-Piret方程建立了恒定pH7分批发酵和补料分批发酵的菌体生长模型和PG合成模型。拟合模型参数后,模型可以合理地描述恒定pH7分批发酵和补料分批发酵的过程。     

MATHEMATICAL MODEL OF L-LACTIC ACID FERMENTATION IN A RDC COUPLED WITH PRODUCT SEPARATION BY ION EXCHANGE

A Rotating Disk Contactor (RDC) was designed to perform the L-lactic acid fermentation with a filamentous fungus, Rhizopus oryzae, which was immobilized on the surface of rotating disks. The bioreactor was operated in repeated batch or continuous modes. The growth rate of the fungi was about 1 mm/day perpendicular to the disks' surface. A weak-base anionic resin, D354, was selected which was high in selectivity for lactic acid separation. Even at low concentration, the ion exchange capability was about 0.5 g Lac/g dry resin. A coupled process of L-lactic acid fermentation and ion-exchange separation was evaluated experimentally. The results indicated that the pH value of the fermentation broth could be maintained at pH 3.5 without any addition of alkali. The conversion ratio of glucose to L-lactic acid was about 0.7 g/g and the fermentation rate was able to reach as high as 62.5g glucose per hour per square meter of the disk surface area. A mathematical model was proposed to describe the simultaneous process of L-lactic acid fermentation and separation by ion exchange, in which the thickness increase of mycelia as well as the substrate and product inhibitions were included. The model simulation was in good agreement with the experimental data.     

木素过氧化物酶降解偶氮染料卡布龙红的动力学及过程模型

研究了用含有木素过氧化物酶活力的酶液降解偶氮染料卡布龙红的动力学和过程 .在实验基础上 ,提出木素过氧化物酶降解卡布龙红的反应机理 ,并推导出了降解卡布龙红的酶反应速率方程 .应用该速率方程 ,考虑H₂ O₂ 的自消失和H₂ O₂ 对酶失活的影响以及产物卡布龙红正离子自由基的反应 ,建立了酶液分批降解卡布龙红的过程模型 ,模型计算值与实验数据符合良好