Search Results

Modern methods of engine development use complex mathematical models. Adding advanced components such as variable valve trains or direct injection systems to the model increases the degrees of freedom resulting in a high number of measurements for validation. Steadily rising costs for development, time and staff make it crucial for industry to improve the quality of measurements with advanced analysis techniques. Often, such models consider the simulated system as stationary, implying that system variables no longer change with time. This paper presents an internal combustion engine measurement system utilizing algorithms for the real-time evaluation of the state of the engine or its components. Several approaches have been reviewed and tested regarding their applicability. The most straightforward algorithms compare the gradient of a sensor signal to a pre-defined threshold.

Laser induced fluorescence (LIF) is used to investigate oil transport mechanisms under real engine conditions. The engine oil is mixed with a dye that can be induced by a laser. The emitted light intensity from the dye correlates with the residual oil at the sensor position and the resulting oil film thicknesses can be precisely determined for each crank angle. However, the general expectation is not always achieved, e.g. an exact representation of piston ring barrel shapes. In order to investigate the responsible lubrication effects of this behavior, a new cavitation algorithm for the Reynolds equation has been developed. The solution retains the mass conservation and does not use any switch function in its mathematical approach. In contrast to common approaches, no vapor-liquid ratio is used, but one or several bigger bubbles are approximated, as have been observed in other experiments already.

Further advancing key technologies requires the optimization of increasingly complex systems with strongly interacting parameters—like efficiency optimization in engine development for optimizing the use of energy. Systematic optimization approaches based on metamodels, so-called Metamodel-Based Design Optimization (MBDO), present one key solution to these demanding problems. Recent advanced methods either focus on Single-Objective Optimization (SoO) on local metamodels with online adaptivity or Multiobjective Optimization (MoO) on global metamodels with only limited adaptivity. In the scope of this work, a fully online adaptive (“in the loop”) optimization approach, capable of both SoO and MoO, is developed which automatically approximates the global system response and determines the (Pareto) optimum.