Survyla Arnas, Levisauskas Donatas, Urniezius Renaldas, Simutis Rimvydas
Department of Automation, Kaunas University of Technology, Studentu 48, LT-51367 Kaunas, Lithuania.
Comput Struct Biotechnol J. 2021 Oct 19;19:5856-5863. doi: 10.1016/j.csbj.2021.10.015. eCollection 2021.
The cell cultivation process in a bioreactor is a high-value manufacturing process that requires excessive monitoring and control compatibility. The specific cell growth rate is a crucial parameter that describes the online quality of the cultivation process. Most methods and algorithms developed for online estimations of the specific growth rate controls in batch and fed-batch microbial cultivation processes rely on biomass growth models. In this paper, we present a soft sensor - a specific growth rate estimator that does not require a particular bioprocess model. The approach for online estimation of the specific growth rate is based on an online measurement of the oxygen uptake rate. The feasibility of the estimator developed in this study was determined in two ways. First, we used numerical simulations on a virtual platform, where the cell culture processes were theoretically modeled. Next, we performed experimental validation based on laboratory-scale (7, 12, 15 L) bioreactor experiments with three different Escherichia coli BL21 cell strains.
生物反应器中的细胞培养过程是一个高价值的制造过程，需要高度的监测和控制兼容性。比生长速率是描述培养过程在线质量的关键参数。大多数为分批和补料分批微生物培养过程中比生长速率的在线估计而开发的方法和算法都依赖于生物量生长模型。在本文中，我们提出了一种软传感器——一种不需要特定生物过程模型的比生长速率估计器。比生长速率的在线估计方法基于对氧气摄取速率的在线测量。本研究中开发的估计器的可行性通过两种方式确定。首先，我们在虚拟平台上进行了数值模拟，在该平台上对细胞培养过程进行了理论建模。其次，我们基于实验室规模（7、12、15升）的生物反应器实验，使用三种不同的大肠杆菌BL21细胞菌株进行了实验验证。
