Search

This paper introduces a method to simplify a nonlinear problem in order to use linear finite element analysis. This approach improves calculation time by 2 orders of magnitude. It is then possible to optimize the geometry of the components even without supercomputers. In this paper the method is applied to a very critical component: the aluminium alloy piston of a modern common rail diesel engine. The method consists in the subdivision of the component, in this case the piston, in several volumes, that have approximately a constant temperature. These volumes are then assembled through congruence constraints. To each volume a proper material is then assigned. It is assumed that material behaviour depends on average temperature, load magnitude and load gradient. This assumption is valid since temperatures vary slowly when compared to pressure (load). In fact pressures propagate with the speed of sound. The method is validated by direct comparison with nonlinear simulation of the same component, the piston, taken as an example. In general, experimental tests have confirmed the cost-effectiveness of this approach. doi:10.14456/WJST.2015.12

A significant issue in aircraft engines is quantifying residual life to overhaul. The algorithm described in this paper calculates with a good level of reliability the residual life of a petrol piston engine. The method was tested on small, latest-generation, naturally-aspirated aircraft and racing piston engines, and has been effective in several experiments. This method is implemented directly on the electronic control system of the engine with very few lines of C-code. The method can also be used in many industrial engines. This innovative method assumes that only two main factors (power level and wear) affect engine durability or time between overhauls. These two factors are considered as separate and combined with worst case criteria. The wear is assumed to follow a logarithmic law and a formula similar to the Miner’s law for material fatigue is used, making it possible to calculate the power-level curve with knowledge of only two points. The wear-curve is also related to elapsed engine cycles. The algorithm is very simple and can be implemented with just a few lines of software code accessing data collected from existing sensors. The system is currently used to evaluate actual residual life of racing engines.