B3.2 Adaptive Soft Sensor for Bioprocess Monitoring
- SMSI 2020
(did not take place because of Covid-19 virus pandemic)
- SMSI 2020 - Sensors and Instrumentation
- B3 Biosensors and Sensors for Biology
- M. Siegl, V. Brunner, D. Geier, T. Becker - Technische Universität München, Freising (Germany)
- 113 - 114
Soft sensors can be used to predict variables that cannot be measured directly. However, even these soft sensors are subject to errors that reduce the accuracy of the prediction. One way to overcome this is to predict a target quantity redundantly using independent measurement systems for the input vari-ables. This study reports on the development of an algorithmic system for combining the redundant submodels to one reliable soft sensor. The proof of concept was conducted with a Pichia pastoris bio-process.