Modeling and Control of a Hybrid Opposed Piston Engine

Abstract

This paper presents the modeling and control of an opposed piston (OP) engine in a novel hybrid architecture. The OP engine was selected for this work due to the inherent thermody-namic benefits and the balanced nature of the engine. The typical geartrain required on an OP engine was exchanged for two electric motors, significantly reducing friction and decoupling the crankshafts. By using the motors to control the crankshaft motion profiles, this configuration introduces capabilities to dynamically vary compression ratio, combustion volume, and scavenging dynamics. To realize these opportunities, a model of the system capturing the instantaneous engine dynamics is essential along with methodology to regulate the crankshaft’s rotational dynamics utilizing the electric motors. The modeling presented here couples a 1D model capturing the gas exchange process during scavenging and a 0D model of the crankshaft dynamics and the heat release profile due to combustion. With the use of this model, a linear quadratic controller with reference feedforward was designed to track the crankshaft motion trajectory. Experimental results are used to validate the model and controller performance. These results highlight the sensitivity to model uncertainty at points with high cylinder pressure, leading to large differences in control input near minimum volume. The proposed controller is, however, still able to maintain tracking error for crankshaft position below ± 1 degree.

Optimizing the fuel efficiency of an opposed piston engine for electric power generation

“…The resulting finite dimensional optimization problem is solved using IPOPT. 23 The system dynamics referred to in equation (4a), fully defined in the Appendix A, are a simplified version of the modeling described in the high-fidelity model presented previously 21 in regards to the coupling between engine crankshaft and motor as well as the open cycle breathing and thermodynamic models. However, despite this simplification, discrete changes are still present in the system dynamics due to the intake and exhaust port actuation, fuel injection, and the combustion process.…”

“…A description of the high-fidelity co-simulation an be found in elsewhere. 21 Iterations between the high-fidelity modeling and the optimization process are repeated until the pressure and temperature of the high-fidelity model at the port closing event are within 0.01 bar and 0.1 K, respectively, of the previous iteration values. Further, the peak cylinder pressure in the optimization model must be within 0.1 bar of the previous cycle, showing convergence of the optimization process.…”

Identification and analysis of the factors to realizing variable intake/exhaust phases in opposed-piston two-stroke diesel engines

“…The port width only affects the area of the intake and exhaust ports; the port height not only affects the port area but also affects the intake and exhaust phases. 17–19 The port inclination contains radial inclination and axial inclination. The radial inclination generates a swirling flow, 20,21 while the axial inclination angle creates a tumbling flow.…”

Trapped equivalence ratio determination in two-stroke engines

“…Resultantly, there has been a resurgence of interest in automotive two-stroke engine research in recent times. 8,11–17…”

“…The computed TER and temporal insights gained regarding the flow physics of scavenging will be used to develop improved simple gas-exchange models that are less reliant on engine/operating-condition-specific empirical tuning, as is the case for currently available models, 18–22 and are more phenomenological in nature. Such models will help develop robust and flexible control systems for managing engine operation across an increasingly wide-ranging list of control variables, for example, fuel composition variations, 23 variable valve/port opening timing, 12,16 variable compression ratio, 8 multiple injections. 11,17 Moreover, such phenomenological models can be integrated into thermodynamic 1D engine models to help in the engine design and development stages.…”