Improving Microalgae Growth Modeling of Outdoor Cultivation with Light History Data using Machine Learning Models: A Comparative Study

Publikation: Beitrag in FachzeitschriftForschungsartikelBeigetragenBegutachtung

Beitragende

  • Tehreem Syed - , Professur für Prozessleittechnik (Erstautor:in)
  • Jonathan Mädler - , Arbeitsgruppe Systemverfahrenstechnik (Autor:in)
  • Leon Urbas - , Professur für Prozessleittechnik (Autor:in)
  • Yen-Cheng Yeh - , Fraunhofer Institute for Interfacial Engineering and Biotechnology IGB, Nobelstraße 12, 70569 Stuttgart, Germany (Erstautor:in)
  • Gordon Brinitzer - , Fraunhofer Institute for Interfacial Engineering and Biotechnology IGB, Nobelstraße 12, 70569 Stuttgart, Germany (Autor:in)
  • Konstantin Frick - , Fraunhofer Institute for Interfacial Engineering and Biotechnology IGB, Nobelstraße 12, 70569 Stuttgart, Germany (Autor:in)
  • Ulrike Schmid-Staiger - , Fraunhofer Institute for Interfacial Engineering and Biotechnology IGB, Nobelstraße 12, 70569 Stuttgart, Germany (Autor:in)
  • Bernard Haasdonk - , Universität Stuttgart (Autor:in)
  • Günter E.M. Tovar - , Fraunhofer Institute for Interfacial Engineering and Biotechnology IGB, Nobelstraße 12, 70569 Stuttgart, Germany (Autor:in)
  • Felix Krujatz - , Professur für Bioverfahrenstechnik (Autor:in)

Abstract

Accurate prediction of microalgae growth is crucial for understanding the impacts of light dynamics and optimizing production. Although various mathematical models have been proposed, only a few of them have been validated in outdoor cultivation. This study aims to investigate the use of machine learning algorithms in microalgae growth modeling. Outdoor cultivation data of Phaeodactylum tricornutum in flat-panel airlift photobioreactors for 50 days were used to compare the performance of Long Short-Term Memory (LSTM) and Support Vector Regression (SVR) with traditional models, namely Monod and Haldane. The results indicate that the machine learning models outperform the traditional models due to their ability to utilize light history as input. Moreover, the LSTM model shows an excellent ability to describe the light acclimation effect. Last, two potential applications of these models are demonstrated: 1) use as a biomass soft sensor and 2) development of an optimal harvest strategy for outdoor cultivation.

Details

OriginalspracheEnglisch
Aufsatznummer129882
FachzeitschriftBioresource Technology
Jahrgang390
PublikationsstatusVeröffentlicht - 28 Okt. 2023
Peer-Review-StatusJa

Externe IDs

Scopus 85175179591
ORCID /0000-0001-5165-4459/work/148145628
ORCID /0000-0001-7012-5966/work/148145988
Mendeley 48733afd-1787-3113-b344-2a4c70fc044f