Monitoring bioreactors using principal component analysis: production of penicillin G acylase as a case study

Abstract

The complexity of biological processes often makes impractical the development of detailed, structured phenomenological models of the cultivation of microorganisms in bioreactors. In this context, data pre-treatment techniques are useful for bioprocess control and fault detection. Among them, principal component analysis (PCA) plays an important role. This work presents a case study of the application of this technique during real experiments, where the enzyme penicillin G acylase (PGA) was produced by Bacillus megaterium ATCC 14945. PGA hydrolyzes penicillin G to yield 6-aminopenicilanic acid (6-APA) and phenyl acetic acid. 6-APA is used to produce semi-synthetic beta-lactam antibiotics. A static PCA algorithm was implemented for on-line detection of deviations from the desired process behavior. The experiments were carried out in a 2-L bioreactor. Hotteling's T(2) was the discrimination criterion employed in this multivariable problem and the method showed a high sensibility for fault detection in all real cases that were studied.

A novel toolbox for E. coli lysis monitoring

“…Although the PCA score plots can be interpreted to monitor bioprocesses with respect to various PCs [13], multi-dimensional analysis of scores and loadings is cumbersome. Therefore, we used a univariate statistic (Hotelling’s T2) from the PCA model to follow deviations from pre-defined operating conditions in the E. coli bioprocesses [13].…”

“…Therefore, we used a univariate statistic (Hotelling’s T2) from the PCA model to follow deviations from pre-defined operating conditions in the E. coli bioprocesses [13]. …”

Robust artificial intelligence tool for automatic start-up of the supplementary medium feeding in recombinant E. coli cultivations

“…There is a strong dependence on previous process stages (strain selection, inoculum preparation and adaptation), which may randomly affect the bioreactor performance. Data sets are usually multidimensional and complex [3,4].…”

Fault diagnosis of a benchmark fermentation process: a comparative study of feature extraction and classification techniques

“…These methods can handle highly correlated data sets and allow analysis and visualization which aids in the understanding of process data and possibly of the process itself. Furthermore, these MV techniques are well suited for on-line statistical process monitoring (SPM) and FDI of batch production and biotechnological processes [4].…”

“…The penicillin production process has been used by several authors [1][2][3][4] as a case study to address the problem of batch process monitoring and fault detection and identification (FDI) and it is considered a benchmark for batch processes as the Tennessee Eastman process is for continuous processes. There is an extensive literature on the modeling of penicillin production but many of the reported models are too simplified or do not consider the effects on biomass growth and penicillin production of important operating variables, such as temperature, pH, agitation power or substrate feed flow rate.…”