Application of kernel partial least square to predict biological hydrogen production by enterobacter aerogenes

Hydrogen production by dark fermentation, where hydrogenase enzymes catalyze the oxidation or evolution of molecular hydrogen from two protons (H+) and electrons, is an economic and environmentally friendly technology for producing clean energy. However, the long-term operations of a continuous anaerobic reactor for fermentative hydrogen production were frequently unstable. In this study, a kernel partial least squares (KPLS) algorithm is employed to develop an online estimation of the key process variables in a biological hydrogen production process by Enterobacter aerogenes in minimal time and with minimal cost. The KPLS approach is potentially very efficient for predicting key quality variables of nonlinear processes by mapping an original input space into a high-dimensional feature space. The proposed kernel-based algorithm could effectively capture the nonlinear relationship in the process variables and show far better performance in the prediction of the key process variable compared with the conventional linear PLS and other nonlinear PLS methods.