Fitting second-order cone constraints to microbial growth data

Second-order cone programming is a highly tractable convex optimization class. In this paper, we fit general second-order constraints to data. This of use when one must solve large-scale, nonlinear problems, but modeling either impractical or does not lead otherwise constraints. Our motivating application biochemical process optimization, in which seek microbial growth The fitting problem nonconvex. We it using the concave–convex procedure, takes form sequence programs. validate our approach on simulated and experimental data, compare its performance with conventional least-squares fitting.