A Soft Sensor-Based Fault-Tolerant Control on the Air Fuel Ratio of Spark-Ignition Engines

Abstract

The air/fuel ratio (AFR) regulation for spark-ignition (SI) engines has been an essential and challenging control problem for engineers in the automotive industry. The feed-forward and feedback scheme has been investigated in both academic research and industrial application. The aging effect can often cause an AFR sensor fault in the feedback loop, and the AFR control performance will degrade consequently. In this research, a new control scheme on AFR with fault-tolerance is proposed by using an artificial neural network model based on fault detection and compensation, which can provide the satisfactory AFR regulation performance at the stoichiometric value for the combustion process, given a certain level of misreading of the AFR sensor.

Keywords

• spark-ignition (SI) engines
• nonlinear dynamics
• artificial neural networks
• fault-tolerant control
• air/fuel ratio (AFR)