LPV Delay-Dependent Sampled-Data Output-Feedback Control of Fueling in Spark Ignition Engines

TL;DR

This paper introduces a delay-dependent sampled-data LPV control method for regulating air-fuel ratio in spark ignition engines, effectively handling varying delays and sampling times to improve engine performance and emissions.

Contribution

It develops a novel LPV control technique that accounts for time delays and sampling effects, providing a convex LMI-based synthesis for improved AFR regulation in engines.

Findings

Achieves precise AFR tracking in simulations

Demonstrates disturbance attenuation capabilities

Ensures stability with varying delays and sampling times

Abstract

We propose a delay-dependent sampled-data output-feedback LPV control technique to address the air-fuel ratio (AFR) regulation problem in spark ignition (SI) engines. AFR control and advanced fueling strategies are essential for maximizing fuel economy while minimizing harmful exhaust emissions. The fuel path of the SI engine, as well as the three-way catalyst (TWC) simplified dynamics, have been captured by a continuous-time linear parameter-varying (LPV) system with varying time delay, where the system dynamics rely on the engine speed, defined as the system's scheduling parameter. The interconnection of the continuous-time plant and a digital controller through analog-to-digital and digital-to-analog converter devices forms a hybrid closed-loop configuration. Therefore, in order to benefit from continuous-time control synthesis tools, the input-delay method has been employed to transform the hybrid closed-loop system into the continuous-time domain with system inherent time delay and an additional delay imposed by the mapping approach. The designed sampled-data gain scheduled output-feedback controller seeks to establish the closed-loop asymptotic stability and a prescribed level of performance for the LPV system with an arbitrarily varying time delay and varying sampling time, where the synthesis results are provided in a convex linear matrix inequality (LMI) constraint setting. Finally, several closed-loop simulation scenarios are conducted, and comparisons are provided to demonstrate the proposed methodology's performance in achieving precise reference AFR tracking and disturbance attenuation.