Development of Advanced Combustion Strategies for CI Engines

The compression ignition (CI) engine is widely used in heavy-duty as well as light-duty applications due to its high efficiency caused by high compression ratio, short combustion duration, and unthrottled air operation. In conventional CI combustion, high-reactivity fuels are injected close to top dead center (TDC) initiating mixing-controlled combustion. This diffusive combustion can lead to high emissions of nitric oxides (NO_x) and particulate matter (PM), due to high-temperature, slightly lean regions, and very rich areas, respectively. NO_x and PM have negative impact on the environment and can cause human respiratory diseases. Therefore, research efforts have focused on minimization of engine-out emissions as well as operation costs, maximization of overall engine efficiency, and reduction of dependency on exhaust after-treatment devices. In order to achieve these goals,advanced combustion strategies with in-cylinder NO_x and soot reduction methods are required .

One approach to counteract the drawbacks of CI engines is the low-temperature combustion (LTC) concept, which has been proposed by many researchers. The LTC concept has the potential to simultaneously reduce nitric oxides as well as soot, because of lower peak temperatures and increased homogeneity. However, an increase in total unburnt hydrocarbons (THC) and carbon monoxide (CO) is often observed. Applied to a CI engine, this concept is frequently called premixed charge compression ignition (PCCI). It is characterized by relatively early injection timings and high external exhaust gas recirculation (EGR). Unfortunately, this strategy also tends to cause very early combustion phasing (CA50), resulting in high noise and lower engine efficiency. To resolve these issues, further investigations are required. The objective of this study is to investigate and evaluate the impact of different injection strategies (injection timing, injection duration, and amount of injections) on performance parameters. Moreover multiply injections are used in order to control the combustion process directly with modern rapid control prototyping (RCP) methods. Therefore an advanced hardware is required. The engine is controlled by an open ECU (VERA) by VEMAC. This ECU is by-passed by and advanced prototyping ECU the Microautobox (MABX) by dSPACE. Analyse of the causal chain between injection and combustion as well as the chain between combustion and emission are main goals of the actual investigations. For analyzing this connection 0d-Models and artificial neural networks (ANN) can be used.

The experiments were carried out on a modified single-cylinder research engine. It has an overall displacement of 0.390 l, with bore and stroke of 75.0 mm and 88.3 mm, respectively. A high-pressure, common-rail fuel injection system (electronically controlled) with maximum injection pressure of 2000 bar is used for the diesel injection. A centrally located piezo injector (CRI3.20) with eight equally spaced orifices with nominal diameter of 0.115 mm is utilized. An overview of engine and injector specifications is given in Table 1.