A sequential convex moving horizon estimator for bioprocesses

We design moving horizon state estimators for a general model of bioprocesses. The underlying optimization is nonconvex due to the microbial growth kinetics, which are modeled as nonlinear functions. relax constraints so that becomes second-order cone program, can be solved efficiently at large scales. Unfortunately, solutions relaxation inexact and thus lead inaccurate estimates. To recover feasible, albeit potentially locally optimal solutions, we use concave–convex procedure, here takes form sequence programs. find outperform unscented Kalman filter on numerical examples based gradostat anaerobic digestion when there high process noise or parameter error.