Search Results

The presented paper deals with a methodology to model cycle-to-cycle variations (CCV) in 0-D/1-D simulation tools. This is achieved by introducing perturbations of combustion model parameters. To enable that, crank angle resolved data of individual cycles (pressure traces) have to be available for a reasonable number of engine cycles. Either experimental data or 3-D CFD results can be applied. In the presented work, experimental data of a single-cylinder research engine were considered while predicted LES 3-D CFD results will be tested in the future. Different engine operating points were selected - both stable ones (low CCV) and unstable ones (high CCV). The proposed methodology consists of two major steps. First, individual cycle data have to be matched with the 0-D/1-D model, i.e., combustion model parameters are varied to achieve the best possible match of pressure traces - an automated optimization approach is applied to achieve that.

The paper deals with model based predictive control of combustion engines. Nonlinear black-box predictive models based on neuro-fuzzy approach are utilized. The structure of the models is optimized within an identification process. The nonlinear models are locally linearized and consequently used for the efficient on-line computation of forthcoming control actions. In desire to respect a fact that the speed of input-output response may vary significantly for different input/output groups, multilevel predictive models are proposed. Predictive control is again applied to approximate the desired behavior of chosen output variables. Potential algebraical constraints between different prediction layers are involved in the control algorithm using quadratic programming. The control scheme is optimized using simplified fast simulation model.

Transient engine operations are modeled and simulated with a 1-D code (GT Power) using heat release and emission data computed by a 3-D CFD code (Kiva3). During each iteration step of a transient engine simulation, the 1-D code utilizes the 3-D data to interpolate the values for heat release and emissions. The 3-D CFD computations were performed for the compression and combustion stroke of strategically chosen engine operating points considering engine speed, torque and excess air. The 3-D inlet conditions were obtained from the 1-D code, which utilized 3-D heat release data from the previous 1-D unsteady computations. In most cases, only two different sets of 3-D input data are needed to interpolate the transient phase between two engine operating points. This keeps the computation time at a reasonable level. The results are demonstrated on the load response of a generator which is driven by a medium-speed diesel engine.

One-dimensional simulation methods for unsteady (transient) engine operations have been developed and published in previous studies. These 1-D methods utilize heat release and emissions results obtained from 3-D CFD simulations which are stored in a data library. The goal of this study is to improve the 1-D methodology by optimizing the control strategies. Also, additional independent parameters are introduced to extend the 3-D data library, while, as in the previous studies, the number of interpolation points for each parameter remains small. The data points for the 3-D simulations are selected in the vicinity of the expected trajectories obtained from the independent parameter changes, as predicted by the transient 1-D simulations. By this approach, the number of time-consuming 3-D simulations is limited to a reasonable amount.

The paper deals with an application of advanced simulation methods to modeling of hydrogen fueled engines. Two models have been applied - 0-D algorithm and CFD. The 0-D model has been based on GT-Power code. The CFD model has been based on Advanced Multizone Eulerian Model representing general method of finite volume. The influence of main engine parameters, e.g. air excess, spark timing, compression ratio, on NOx formation and engine efficiency has been investigated. Both models have been calibrated with experimental data. Examples of results and comparison with experiments are shown. The means of reducing NOx formation are discussed.

The paper describes patterns of algorithms for different innovative levels of design at parametric, configuration and conceptual levels. They can be applied to Computer-aided Engine Design (CED). Data structures, process simulation hierarchy, used modules of engine simulation and needs for their further development are described. An example of advanced thermodynamics modeling of combustion engines is included.

The paper deals with the simulation of properties of IC engine equipped with a chemically inert porous media (PM) to homogenize and stabilize the combustion of CI engines. The purpose of the PM matrix use is to ensure reliable a ignition of lean mixture and to limit maximum in-cylinder temperature during combustion. It is aimed at NO formation reduction. The influence of PM on an engine cycle is examined by means of CFD simulations. Results demonstrating the influence of heat accumulation, heat supply during compression and expansion strokes and self-ignition properties of a fuel on the engine cycle are presented. All simulations involve modeling of NO formation. The homogenization capability and the flame stabilization one of the PM are discussed.

The presented work deals with possibilities of modeling divided combustion chamber using available 1-D/0-D software. It is usable for indirect injection diesel engines, gas SI engines with pre-chambers for very lean mixture ignition, etc. The model solves all layouts where main cylinder is connected to additional volumes. This connection allows for heat and energy transfer between connected parts. The application of standard ROHR functions (Wiebe, etc.) which are normalized to constant fuel mass is limited. A new marker gas concentration algorithm is proposed for the use of empiric ROHR functions. The standard approach (without proposed algorithm) was tested modeling large-bore gas SI engine with pre-chamber where the mixture is ignited and experimental direct injection hydrogen one-cylinder engine with an additional volume between fuel injector and the cylinder itself to protect the injector from very high pressures and temperatures in the cylinder.

This article presents results of the Design Assistance System (DASY, will be referred to as a tool in this paper) development and examples of its application for engine concept modeling. The software tool for creating and maintaining knowledge database is being developed at the Czech Technical University in Prague. This tool is targeted to simplify and speed up the concept design process. The targets were met by providing the high level of flexibility along with a simple user interface. Two examples that show interaction of this tool with computer-aided design (CAD) and computer-aided engineering (CAE) software are presented. One example includes an optimization using implemented genetic algorithm. By using this tool, one can create templates for conserving the knowledge acquired during engine design in the past. It provides hints for the future design tasks by offering a data of similar designs, based on experiments and simulations at different levels of complexity and profoundness.

The paper describes the tool of ICE transient response simulation suitable for incorporation into a predictive engine controller. The model is simplified, thus enhancing the simulation speed but keeping its predictive capability at a reasonable level. The main modules of a code suitable for the near-real-time simulation of engine thermodynamics are described in the paper. They include engine cylinder (incl. simplified pressure trace prediction), fuel injection system, main controllers, both inlet and exhaust manifolds, turbocharger and engine dynamics. The laws of conservation are used to describe any of the thermodynamic/hydrodynamic modules of a model. The method of algebraic re-construction of a pressure trace inside a cylinder has been developed and tested for prediction of engine speed variation. The modular structure of a model allows for the implementation of the current operating principles of ICEs.

Diesel fuel injection process calculations have been performed by means of in-house developed mathematical models. An Eulerian multidimensional code for in-cylinder two-phase flow computations is used in conjunction with a hydrodynamic one-dimensional model of a fuel injection system. The multidimensional model comprehends all basic processes, which play a role in spray formation. The compressible gaseous flow with transport of species is solved together with the flow of dispersed liquid phase using the Eulerian reference frame for both phases. The two-way coupling between the phases in mass, momentum, and energy balances is considered. A detailed description of liquid phase is present, taking into account drop size distribution in terms of the multi-continua approach. The hydrodynamic model capable of simulating common fuel injection systems is used for the rate-of-injection computations to provide realistic boundary conditions to the spray model.

In this article the development of a combustion system with a fuel-scavenged pre-chamber is described. Such a system is commonly used in large-bore engines operated with extremely lean mixtures. The authors implemented the scavenged pre-chamber into a light duty truck-size engine with a bore of 102 mm. The lean burn strategy is intended to achieve very low nitrogen oxide (NOx) emissions at low load. At full load a stoichiometric mixture strategy is applied to achieve sufficient power density while simultaneously enabling the use of a relatively simple three-way catalytic converter for exhaust gas aftertreatment. This work outlines the pre-chamber design features and introduces the results of an experimental investigation of the effect of pre-chamber ignition on a single cylinder testing engine.

The article describes the utilization of the map-less approach in simulation of single and twin scroll radial turbines. The conventional steady flow maps are not used. An unsteady 1-D model of a twin scroll turbine includes scrolls, mixing of flows upstream of the impeller, turbine wheel, leakages and outlet pipe. Developed physical turbine model was calibrated with data from experiments at specific steady flow turbocharger test bed with open loop, which enables to achieve arbitrary level of an impeller admission via throttling in separate sections. A selected twin scroll turbine was tested under full, partial flow admission of an impeller and extreme partial admission with closed section. The required number of operating points is relatively low compared with conventional steady flow maps, when the maps have to be generated for each level of an impeller admission. The calibration process of the full 1-D turbine model is described.

The paper describes a way to a 1-D central streamline model of a radial turbine flow, suitable for twin-scroll description and based on approximation of real physics of flow mixing and energy transformation. The original 1-D model of a single scroll turbine, described earlier in numerous SAE papers, has been amended by twin-scroll nozzles (both vaneless or with blade cascades) and mixing of individual partitions of flows upstream of additional vaneless nozzle and an impeller. This model is transferable to 1-D unsteady simulations as it is (i.e., using quasi-steady approach) or using 1-D unsteady solvers. It has suitable features even for more detailed description of turbine flows and energy transformation. The first results of pulse influence on turbine maps delivered expected results consisting of complicated interaction between individual losses.

Current vehicles, especially the electric ones, are complex mechatronic devices. The pickup vehicles of small sizes are currently used in transport considerably. They often operate within a repeating scheme of a limited variety of tracks and larger fleets. Thanks to mechatronic design of vehicles and their components and availability of high capacity data connection with computational centers (clouds), there are many means to optimize their performance, both by planning prior the trip and recalculations during the route. Although many aspects of this opportunity were already addressed, the paper shows an approach developed to further increase the range of e-vehicle operation. It is based on prior information about the route profile, traffic density, road conditions, past behaviour, mathematical models of the route, vehicle and dynamic optimization. The most important part of the procedure is performed in the cloud, using both computational power and rich information resources.