Effects of Intake Pressure and Air-Fuel Ratio on Controlled Hot Surface Ignition Operating with Natural Gas 2015-32-0721

In an effort to reduce both maintenance costs and NO_x emissions of small cogeneration engines operated with natural gas, an alternative ignition system that allows stable operation at very lean homogeneous air-fuel mixtures has been developed. Combustion is induced by an electrically heated ceramic glow plug, whose temperature is controlled by an ECU. Adjusting hot surface temperature allows shifting the inflammation timing of the mixture and, therefore, the phasing of combustion in the engine cycle.

The main aim of this work was to determine the effect of intake pressure and air-fuel ratio on the parameters of hot surface ignition (HSI) and understand which are the factors limiting stable HSI operation in terms of cycle-by-cycle variations. Furthermore, in order to explain abnormal combustion phenomena occurring at high surface temperatures, the process of mixture inflammation and combustion was examined through a combination of numerical simulation using the ECFM-3z combustion model and engine trials employing optical probes to record combustion radiation.

The experiments showed that the energy consumption required to heat the surface and initiate combustion can be decreased substantially when raising intake pressure and/or advancing combustion phasing. On the other hand, when leaning the mixture or retarding combustion, the energy demand increases. It was found that the temperature control of the hot surface is most stable when heat released from combustion is relatively low and produces low combustion temperatures. When this is not the case, e.g. at lower air-fuel ratios or advanced combustion phasings, cyclic variability in heat release rate and phasing have a strong impact on surface temperature and it becomes more difficult for the HSI controller to keep hot surface temperature constant. As a result, the timing of mixture inflammation fluctuates considerably, leading to high cyclic dispersion in CA50, IMEP and p_max.