Ion-Current Sensor based Closed-Loop Control of Lean Gasoline Combustion with High Compression Ratio

点击翻译按钮获取中文摘要

Lean combustion offers a high potential for improvement of the energy efficiency of gasoline engines, particularly in combination with a further reduction of throttle losses and higher compression ratios. Homogenous Charge Compression Ignition (HCCI) combustion can reduce NOx raw emission by up to 99 %, when compared to conventional spark-ignition (SI) lean combustion.Generally speaking, lean combustion is very sensitive to changing global and local parameters in the combustion chamber that cannot be measured directly. Accordingly, lean combustion has major drawbacks in controllability, especially if the HCCI combustion mode is used. Fluctuations of the cylinders charge state cause cyclic deviations of the combustion with strong effects on efficiency and emissions and process disturbances up to misfiring. This underlines the necessity of fast in-cycle closed-loop control algorithms to actively stabilize the process. Research hypothesis of the applicants is, that signals from an ion current sensor can deliver additional information about the state of the cylinder charge that can improve the controllability of lean combustion. The effects of formation of intermediate species on the ion current require new approaches on the hard-and software side and have not yet been analyzed in detail. Additionally, the hardware analysis circuits that are used for interpretation of the ion current sensor signal have to be improved significantly. A systematic methodology is required to identify the optimum electrical sensor layout and the software signal processing algorithm.For SI combustion, predominantly with high compression ratios, pre-ignition of the cylinder charge is a challenging problem. For active prevention of this phenomena, a fast and reliable identification of an approaching pre-ignition is required. Advanced analysis of weak ion current signals is assumed to deliver additional information about the cylinder charge and allows to establish new FPGA-based in-cycle control interventions.Analysis in the frequency domain with digital signal processors is a promising approach for utilization in fast in-cycle control algorithms to stabilize lean combustion. In this project, the applicants target a deeper understanding about the correlations of the ion current sensors signal and the underlying chemical and physical effects in the cylinder charge and the resulting conductivity. To improve the measurement and signal preprocessing methodology, a detailed simulation is combined with investigations on test engines in Shanghai and Aachen. The ion current analysis hardware setup is adapted to improve the signal to noise ratio. The identified correlation between the ion current and cylinder charge state will be used to perform a feasibility study for new FPGA-based in-cycle control algorithms at the test bench.