Shape-Constrained Splines as Transparent Black-Box Models for Bioprocess
Identification of mathematical models is an important task for the design and optimization of biokinetic processes. The Monod rate law is often chosen by default, although this rate law is restrictive and cannot capture all biokinetic process dynamics, which ultimately reduces the predictive capability of the resulting models. This paper proposes an alternative rate-law structure consisting of a flexible black-box spline function that is forced to obey a predefined shape. This way, the difficult task of searching through potentially incomplete rate-law libraries can be circumvented. A simulated case study is used to illustrate the applicability of the method and its superiority to represent unconventional growth conditions, where neither Monod nor Tessier kinetics are appropriate.
Record created on 2016-07-22, modified on 2016-12-05