Search Results

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 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.

The objective of this paper is to present the results of the GT Power calibration with engine test results of the air loop system technology down selection described in the SAE Paper No. 2012-01-0831. Two specific boosting systems were identified as the preferred path forward: (1) Super-turbo with two speed Roots type supercharger, (2) Super-turbo with centrifugal mechanical compressor and CVT transmission both downstream a Fixed Geometry Turbine. The initial performance validation of the boosting hardware in the gas stand and the calibration of the GT Power model developed is described. The calibration leverages data coming from the tests on a 2 cylinder 2-stroke 0.73L diesel engine. The initial flow bench results suggested the need for a revision of the turbo matching due to the big gap in performance between predicted maps and real data. This activity was performed using Honeywell turbocharger solutions spacing from fixed geometry waste gate to variable nozzle turbo (VNT).

The aim of the current contribution is to develop a tool for the improvement of accuracy of turbocharger turbine simulation during matching of a turbocharger to an engine. The paper demonstrates the possibility of unsteady turbine simulation in pulsating flow caused by an internal combustion engine using the basic modules of generalized 1-D manifold solver with entities (pipes, channels) under centrifugal acceleration in general direction and under non-uniform angular speed, which has not yet been explored. The developed model extrapolates steady operation turbine maps by this way. It uses 1-D model parameters identified from steady flow experiments. Unlike the lumped-parameter standard models of turbocharger turbines, the model takes into account complete 1-D features of a turbine flow path including arbitrary shape of turbine impeller vanes.

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.

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.

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 the investigation of turbocharger optimization procedures using amended 1-D simulation tools. The proposed method uses scaled flow rate/effficiency maps for different sizes of a radial turbine together with a fictitious compressor map. The compressor pressure ratio/efficiency map depends on compressor circumference velocity only and predicts the both compressor specific power and achievable efficiency. At the first stage of optimization, it avoids the problems of reaching choking/surge limits. It enables the designer to find a suitable turbine type under realistic unsteady conditions (pressure pulses in exhaust manifold) concerning turbine flow area. Once the optimization of turbine/compressor impeller diameters is finished, the specific compressor map is selected. The proposed method provides the fast way to the best solution even for the case of a VGT turbine. Additional features have been developed for the representation of scaled turbine and compressor maps.

The paper deals with the investigation of pressure, flow and temperature sensor performance under unsteady conditions using advanced 1-D codes for simulation of engine operation. Approach of internal combustion engine (ICE) sensor modeling in an engine simulation code is described. Some new external modules have been developed to couple engine-and-pipe model to sensors. Sensor dynamic and engine dynamic effects are separated by combining a sensor model with an engine model. The models were tuned to match real data with the goal of uncovering the transfer function between the measured signal and the actual signal. Procedure for estimation of the in-cylinder pressure pattern from distorted pattern at sensor location using empirical transfer function is presented. The developed model seems to have a wide application, e.g. for investigation of dynamical characteristics of lambda sensors or gas analyzer probes.

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 investigation of flow characteristics of turbocharger turbine under real operating conditions on engine by means of combination of experimental data and advanced 1-D code for combustion engine simulation. Coupling simulations tools with the results of measurements provides the engineers with data which are difficult or impossible to measure. For instance by means of a three pressure analysis (TPA) applicable on engine cylinder the engineers can obtain burn rate, valve flow and residual gas compound from measured pressure traces in cylinder and at inlet and outlet ports. A method for turbocharger turbine on engine identification similar in principle to the three pressure analysis has been applied on radial turbine with variable geometry. A new computational module has been developed to allow identification of instantaneous flow and efficiency characteristics of the turbine.

The new simulation tool consists of an iterative loop of a 3-D code in parallel to a 1-D code that is employed to simulate transient engine cycles. The 1-D code yields the basic pattern of initial and boundary conditions and the 3-D simulations at several typical engine operating points are used to crosscheck the performance as well as aid in the model calibration. A flexible regression model of the fuel burn rate and the associated ROHR has been developed in conjunction with the 3-D simulations using a combination of three added Vibe functions. The emissions at the end of the expansion stroke are also predicted. The parameters of the Vibe functions and emissions are found via nonlinear regression based on state parameters such as engine speed, relative A/F ratio, EGR/rest gas contents, injection timings, etc. Additional 3-D simulations that are made at specific engine operating points complement this compact burn rate parameter library.

The boost pressure demands call for high efficiency turbochargers. Perfect matching to an engine and controlling in operation is a prerequisite, especially if highly diluted mixture is used. The main impact on four-stroke engine efficiency is performed via gas exchange work, Correct turbocharger representation, usually performed by maps, should be delivered by turbocharger manufacturers and applied in simulation optimizations. The robust calibration methods are required for fast-running real time simulations used for model-based control. The paper clarifies the relations between apparent and real turbocharger isentropic efficiencies at steady-flow testbed and their impact on engine cycle optimization by simulation. Simple procedures excluding the impact of heat transfer inside a turbocharger are described. The described methods are based on the use of overall turbocharger efficiency.

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.

A previously developed CFD-based optimization tool is utilized to find optimal engine operating conditions with respect to fuel consumption and emissions. The optimization algorithm employed is based on the steepest descent method where an adaptive cost function is minimized along each line search using an effective backtracking strategy. The adaptive cost function is based on the penalty method, where the penalty coefficient is increased after every line search. The parameter space is normalized and, thus, the optimization occurs over the unit cube in higher-dimensional space. The application of this optimization tool is demonstrated for the Sulzer S20, a central-injection, non-road DI diesel engine. The optimization parameters are the start of injection of the two pulses of a split injection system, the duration of each pulse, the exhaust gas recirculation rate, the boost pressure and the compression ratio.

The objective of this study is the development of a computationally efficient CFD-based tool for finding optimal engine operating conditions with respect to fuel consumption and emissions. The optimization algorithm employed is based on the steepest descent method where an adaptive cost function is minimized along each line search using an effective backtracking strategy. The adaptive cost function is based on the penalty method, where the penalty coefficient is increased after every line search. The parameter space is normalized and, thus, the optimization occurs over the unit cube in higher-dimensional space. The application of this optimization tool is demonstrated for the Sulzer S20, a central-injection, non-road DI diesel engine. The optimization parameters are the start of injection of the two pulses, the duration of each pulse, the duration of the dwell, the exhaust gas recirculation rate and the boost pressure.

The mass and overall dimensions of massively downsized engines for very high bmep (up to 35 bar) cannot be estimated by scaling of designs already available. Simulation methods coupling different levels of method profoundness, as 1-D methods, e.g., GT Suite/GT Power with in-house codes for engine mechanical efficiency assessment and preliminary design of boosting devices (a virtual compressor and a turbine), were used together with optimization codes based on genetic algorithms. Simultaneously, the impact of optimum cycle on cranktrain components dimensions (especially cylinder bore spacing), mass and inertia force loads were estimated since the results were systematically stored and analyzed in Design Assistance System DASY, developed by the authors for purposes of early-stage conceptual design. General thermodynamic cycles were defined by limiting parameters (bmep, burning duration, engine speed and turbocharger efficiency only).

The physical 1-D model of a radial turbine consists in a set of gas ducts featuring total pressure and/or temperature changes and losses. This model has been developed using the basic modules of generalized 1-D manifold solver. The tools for it were presented at SAE 2008 and 2009 World Congresses. The model published before is amended by a semi-empiric mechanical loss and windage loss modules. The instantaneous power of a turbine is integrated along the rotating impeller channel using Euler turbine theorem, which respects the local unsteadiness of mass flow rate along the channel. The main aim of the current contribution is to demonstrate the use of measured turbine maps for calibration of unsteady turbine model for different lay-outs of turbine blade cascades. It is important for VG turbines for the optimal matching to different engine speeds and loads requirements.

This paper introduces a research work on the air loop system for a downsized two-stroke two-cylinder diesel engine conducted in framework of the European project dealing with the POWERtrain for Future Light-duty vehicles - POWERFUL. The main objective was to determine requirements on the air management including the engine intake and exhaust system, boosting devices and the EGR system and to select the best possible technical solution. With respect to the power target of 45 kW and scavenging demands of the two-cylinder two-stroke engine with a displacement of 0.73 l, a two-stage boosting architecture was required. Further, to allow engine scavenging at any operation, supercharger had to be integrated in the air loop. Various air loop system layouts and concepts were assessed based on the 1-D steady state simulation at full and part load with respect to the fuel consumption.

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.