Modeling and predictive control of Free Piston Engine Generator

This paper focuses on the modeling and control of a Free Piston Engine Generator (FPEG) which is an alternative to conventional internal combustion engines to be used as an auxiliary power unit (APU) in hybrid electric vehicles (HEV). A physics-based, continuous-time model and a discrete-time, control-oriented model are derived. A model predictive control approach is proposed to manage system constraints and to control piston position by regulating the fuel injection quantity and external electrical load. Top Dead Center (TDC) clearance height is used for feedback and is estimated using Newton's method and an Extended State Observer (ESO). Simulation results are reported and show that the proposed control framework successfully tracks the target piston clearance height and reduces clearance fluctuations caused by instantaneous electrical load changes.