Search Results

The V-Cycle is a well accepted and commonly implemented process model for systems engineering. The concept phase is represented by the upper-left portion of the V, in which very high level system simulations are the predominant modeling activity. Traveling down the V toward the vertex, sub-system level and component level simulations are employed as one enters the development phase. Finally, the test and validation phase is completed, and is represented by the right side of the V. Simulation tools have historically been used throughout some phases of the V-cycle, and with the ever increasing computing power, and the increasingly accurate and predictive simulation tools available to the engineer, today it is common that simulation is used in every phase of the cycle, from concept straight through the test and validation phases.

As engines advance toward greater efficiency and lower emissions, there is increasing need for accurate real-time residual models for engine control. Both the formulation of real-time-capable models and the development of methods for measuring or estimating residuals during engine calibration have been difficult and longstanding problems. This paper describes development of a low-cost, easy-to-use tool for on-line residual estimation in all cylinders of an IC engine. The basic method, hardware required, and software structure are described. The residual estimation tool was applied to estimate residuals over the operating map in all cylinders of a six-cylinder direct-injection SI engine equipped with dual-independent phasers. The data was used to calibrate a real-time residual model integrated into the engine management system. Validation data confirming accuracy of the model are presented.

This paper presents a novel approach in real-time engine modeling. Unlike standard practices, which involve system level modeling, the presented methodology is a hybrid physical/system domain solution. Specifically, for each subsystem that the engine is divided into, a physical, map-based, or combination physical/map-based solution is chosen depending on the available computational power and the desired model detail. The resulting semi-physical engine models are suitable for real-time applications, such as Hardware-In-Loop (HiL) simulations, and, at the same time, re-usable to a large extent when model updates are required. In addition, since the proposed methodology allows for variable level of detail -from models as simple as pure map-based look-ups for torque, airflow, and exhaust temperature, all the way to models capable of predicting crank angle resolved cylinder pressure- it provides natural adjustability to the ongoing growth of computer power.

Genetic Algorithms have proved to be very useful as global search methods for multi-dimensional optimization problems. One drawback, however, is that they are inefficient from the point of view of the number of function evaluations. This paper presents a two phase approach to optimization, using Game Theory in an initial step which provides a family of designs which are close to the Pareto frontier. The starting population for the genetic algorithm is then selected from the non-dominated designs produced in the first phase. This ensures that the genetic algorithm starts with a population of points which are already optimized to a large degree.

A structured, semi-automatic method for reducing a high-fidelity engine model to a fast running one has been developed. The principle of this method rests on the fact that, under certain assumptions, the computationally expensive components of the simulation can be substituted with simpler ones. Thus, the computation speed increases substantially while the physical representation of the engine is retained to a large extent. The resulting model is not only suitable for fast running simulations, but also usable and updatable in later stages of the development process. The thrust of the method is that the calibration of the fast running components is achieved by use of automatically selected neural networks. Two illustrative examples demonstrate the methodology. The results show that the methodology achieves substantial increase in computation speed and satisfactory accuracy.

Exhaust acoustics simulation is an important part of the exhaust system process. Especially important is the trend towards a coupled approach to performance and acoustics design. The present paper describes a new simulation tool developed for such coupled simulations. This tool is based on a one-dimensional fluid dynamics solution of the flow in the engine manifolds and exhaust and intake elements. To represent the often complex geometries of mufflers, an easy-to-use graphical pre-processor is provided, with which the user builds a model representation of mufflers using a library of basic elements. A comparison made to two engines equipped with exhaust silencers, shows that the predictions give good results.

An increasing emphasis is being placed in the vehicle development process on transient operation of engines and vehicles, and of engine/vehicle integration, because of their importance to fuel economy and emissions. Simulations play a large role in this process, complementing the more usual test-oriented hardware development process. This has fueled the development and continued evolution of advanced engine and powertrain simulation tools which can be utilized for this purpose. This paper describes a new tool developed for applications to transient engine and powertrain design and optimization. It contains a detailed engine simulation, specifically focused on transient engine processes, which includes detailed models of engine breathing (with turbocharging), combustion, emissions and thermal warm-up of components. Further, it contains a powertrain and vehicle dynamic simulation.