On the performances of noise filters in the restoration of oscillatory behavior in continuous yeast cultures.

Continuous flow microbial fermentations under industrial conditions are subject to the influx of noise, mainly through the feed stream. Noise upsets the normal deterministic behavior. For continuous cultures of Saccharomyces cerevisiae exhibiting oscillatory responses, four kinds of commonly used noise filters, three algorithmic and one neural, have been compared for their ability to restore noise-free oscillations. An auto-associative neural filter was the best, similar to earlier observations for other organisms under non-oscillatory conditions. This enhances the general applicability of neural filters for industrial scale fermentations.