Subspace LPV Identification of Secondary Reforming Process

Abstract

LPV identification approach has quite often raised. In general, LPV identification method can be grouped in two forms: input-output and state-space. In previous research, Recursive Least Square algorithm (RLS) is used to get input-output form. The use of other algorithm and form as the development, applied in this research. The algorithm used in this research is the subspace LPV algorithm using MOESP. The resulting model is in state space form. The modeled system is a secondary reforming section on PT Pupuk Kalimantan Timur’s ammonia plant. Sub-process modeled in this paper was a continuation of the subprocesses in previous research, which is a unit complex process.