Search Results

Modeling thermal systems in Battery Electric Vehicles (BEVs) is crucial for enhancing energy efficiency through predictive control strategies, thereby extending vehicle range. A major obstacle in this modeling is the often limited availability of detailed system information. This research introduces a methodology using neural networks for system identification, a powerful technique capable of approximating the physical behavior of thermal systems with minimal data requirements. By employing black-box models, this approach supports the creation of optimization-based operational strategies, such as Model Predictive Control (MPC) and Reinforcement Learning-based Control (RL). The system identification process is executed using MATLAB Simulink, with virtual training data produced by validated Simulink models to establish the method's feasibility. The neural networks utilized for system identification are implemented in MATLAB code.

The calibration of Engine Control Units (ECUs) for road vehicles is challenged by stringent legal and environmental regulations, coupled with short development cycles. The growing number of vehicle variants, although sharing similar engines and control algorithms, requires different calibrations. Additionally, modern engines feature increasingly number of adjustment variables, along with complex parallel and nested conditions within the software, demanding a significant amount of measurement data during development. The current state-of-the-art (White Box) model-based ECU calibration proves effective but involves considerable effort for model construction and validation. This is often hindered by limited function documentation, available measurements, and hardware representation capabilities. This article introduces a model-based calibration approach using Neural Networks (Black Box) for two distinct ECU functional structures with minimal software documentation.

Noise, vibration and harshness (NVH) is one of the most important performance evaluation aspect of electric motors. Among the different causes of the NVH issues of electrical drives, the high-frequency spatial and temporal harmonics of the electrical drive system is of great importance. To reduce the tonal noise of the electric motors, harmonic injection methods can be applied. However, a lot of the existing related work focuses more on improving the optimization process of the parameter settings of the injected current/flux/voltage, which are usually limited to some specific working conditions. The applicability and effectivity of the algorithm to the whole frequency/speed range are not investigated. In this paper, a multi-domain pipeline of harmonic injection controller design for a permanent magnet synchronous motor (PMSM) is proposed.

Numerical analyses of the liquid fuel injection and subsequent fuel-air mixing for a high-tumble direct injection engine with an active pre-chamber ignition system at operation conditions of 2000 RPM are presented. The Navier-Stokes equations for compressible in-cylinder flow are solved numerically using a hierarchical Cartesian mesh based finite-volume method. To determine the fuel vapor before ignition large-eddy flow simulations are two-way coupled with the spray droplets in a Lagrangian Particle Tracking (LPT) formulation. The combined hierarchical Cartesian mesh ensures efficient usage of high performance computing systems through solution adaptive refinement and dynamic load balancing. Computational meshes with approximately 170 million cells and 1.0 million spray parcels are used for the simulations.

Prospective combustion engine applications require the highest possible energy conversion efficiencies for environmental and economic sustainability. For conventional Spark-Ignition (SI) engines, the quasi-hemispherical flame propagation combustion method can only be significantly optimized in combination with high excess air dilution or increased combustion speed. However, with increasing excess air dilution, this is difficult due to decreasing flame speeds and flammability limits. Pre-Chamber (PC) initiated jet ignition combustion systems significantly shift the flammability and flame stability limits towards higher dilution areas due to high levels of introduced turbulence and a significantly increased flame area in early combustion stages, leading to considerably increased combustion speeds and high efficiencies. By now, vehicle implementations of PC-initiated combustion systems remain niche applications, especially in combination with lean mixtures.

Modeling combustion of transportation fuels remains a difficult task due to the extremely large number of species constituting commercial gasoline and diesel. However, for this purpose, multi-component surrogate fuel models with a reduced number of key species and dedicated reaction subsets can be used to reproduce the physical and chemical traits of diesel and gasoline, also allowing to perform CFD calculations. Recently, a detailed surrogate fuel kinetic model, named C3 mechanism, was developed by merging high-fidelity sub-mechanisms from different research groups, i.e. C0-C4 chemistry (NUI Galway), linear C6-C7 and iso-octane chemistry (Lawrence Livermore National Laboratory), and monocyclic aromatic hydrocarbons (MAHs) and polycyclic aromatic hydrocarbons (PAHs) (ITV-RWTH Aachen and CRECK modelling Lab-Politecnico di Milano).

The most challenging part of the engine combustion development is the reduction of pollutants (e.g. CO, THC, NOx, soot, etc.) and CO2 emissions. In order to achieve this goal, new combustion techniques are required, which enable a clean and efficient combustion. For compression ignition engines, combustion rate shaping, which manipulates the injected fuel mass to control the in-cylinder pressure trace and the combustion rate itself, turned out to be a promising opportunity. One possibility to enable this technology is the usage of specially developed rate shaping injectors, which can control the injection rate continuously. A feasible solution with series injectors is the usage of multiple injections to control the injection rate and, therefore, the combustion rate. For the control of the combustion profile, a detailed injector model is required for predicting the amount of injected fuel. Simplified 0D models can easily predict single injection rates with low deviation.

The increasingly stringent limits on pollutant emissions from internal combustion engine-powered vehicles require the optimization of advanced combustion systems by means of virtual development and simulation tools. Among the gaseous emissions from spark-ignition engines, the unburned hydrocarbon (HC) emissions are the most challenging species to simulate because of the complexity of the multiple physical and chemical mechanisms that contribute to their emission. These mechanisms are mainly three-dimensional (3D) resulting from multi-phase physics - e.g., fuel injection, oil-film layer, etc. - and are difficult to predict even in complex 3D computational fluid-dynamic (CFD) simulations. Phenomenological models describing the relationships between the physical-chemical phenomena are of great interest for the modeling and simplification of such complex mechanisms.

In the field of gasoline powertrain calibration, the challenges are growing due to ever shorter time-to-market requirements and a simultaneous increase in powertrain complexity. In addition, the great variety of vehicle variants requires an increasing number of prototypes for calibration and validation tasks within the framework of the current Real Driving Emissions (RDE) regulations and the expected Post Euro 6 emission standards. Hardware-in-the-Loop (HiL) simulations have been introduced successfully to support the calibration tasks in parallel to the conventional vehicle development activities. The HiL approach enables a more reliable compliance with emission limits and improves the quality of calibrations, while reducing the number of prototype vehicles, test resources and thus overall development costs.

Knock control is one of the most vital functions for safe and fuel-efficient operation of gasoline engines. However, all knock control strategies rely on accurate knock detection to operate the engine close to the optimal set point. Knock detection is usually calibrated on the engine test bench, requiring the engine to run with knocking combustion in a time-consuming multi-stage campaign. Model-based calibration significantly reduces calibration loops on the test bench. However, this method requires a large effort in building and validating the model, which is often limited by the lack of function documentation, available measurements or hardware representation. As the software models are often not available, function structures vary between manufacturers and sub model functions are often documented as black boxes. Hence, using the model-based approach is not always possible.

The growing number of passenger car variants and derivatives in all global markets, their high degree of software differentiability caused by regionally different legislative regulations, as well as pronounced market-specific customer expectations require a continuous optimization of the entire vehicle development process. In addition, ever stricter emission standards lead to a considerable increase in powertrain hardware and control complexity. Also, efforts to achieve market and brand specific multistep adjustable drivability characteristics as unique selling proposition, rapidly extend the scope for calibration and testing tasks during the development of powertrain control units. The resulting extent of interdependencies between the drivability calibration and other development and calibration tasks requires frontloading of development tasks.

The reduction of cycle-to-cycle variations (CCV) is a prerequisite for the development and control of spark-ignition engines with increased efficiency and reduced engine-out emissions. To this end, Large-Eddy Simulations (LES) can improve the understanding of stochastic in-cylinder phenomena during the engine design process, if the employed modeling approach is sufficiently accurate. In this work, an inhouse code has been used to investigate CCV in a direct-injected spark ignition (DISI) engine under fuel-lean conditions with respect to a stoichiometric baseline operating point. It is shown that the crank angle when a characteristic fuel mass fraction is burned, e.g. MFB50, correlates with the equivalence ratio computed as a local average in the vicinity of the spark plug. The lean operating point exhibits significant CCV, which are shown to be correlated also with the in-cylinder subfilter-scale (SFS) kinetic energy.

The efficient operation of powertrain test benches in research and development is strongly influenced by the state of “health” of the functional test object. Hence, the use of Early Damage Detection Systems (EDDS) with Unit Under Test (UUT) monitoring is becoming increasingly popular. An EDDS should primarily avoid total loss of the test object and ensure that damaged parts are not completely destroyed, and can still be inspected. Therefore, any abnormality from the standard test object behavior, such as an exceeding of predefined limits, must be recognized at an early testing time, and must lead to a shutdown of the test bench operation. With sensors mounted on the test object, it is possible to isolate the damage cause in the event of its detection. Advanced EDDS configurations also optimize the predefined limits by learning new shutdown values according to the test object behavior within a very short time.

Physics-based models in a closed-loop feedback control of a premixed charge compression ignition (PCCI) engine can improve the combustion efficiency and potentially reduce harmful NOx and soot emissions. A stand-alone multi-zone combustion model has been proposed in the literature using a physics-based mixing approach. The scalar dissipation rate emerged as the determining parameter in the model for mixing among different zones in the mixture fraction space. However, the calculation of the scalar dissipation rate depends on three approaches: three-dimensional computational fluid dynamics (3-D CFD) combustion simulations based on representative interactive flamelet (RIF) model, tabulation, or an empirical algebraic model of the scalar dissipation rate fitted for the given operating conditions of the engine. While the 3-D CFD approach provides accurate results, it is computationally too expensive to use the multi-zone model in closed-loop control.

In the foreseeable future, the transportation sector will continue to rely on internal combustion engines. Therefore, reduction of engine-out emissions and increase in engine efficiency are important goals to meet future legislative regulations and restricted fuel resources. One viable option, which provides lower peak temperatures and increased mixture homogeneity and thus simultaneously reduces nitric oxide as well as soot, is a low-temperature combustion (LTC) concept. However, this might result in an increase of unburnt hydrocarbon, carbon monoxide, and combustion noise due to early combustion phasing and lower engine efficiency. Various studies show that these drawbacks can be compensated by advanced injection strategies, e.g. by employing multiple injections. The aim of this work is to identify the optimum injection strategy, which enables a wide range of engine operating points in LTC mode with reduced engine-out emissions.

The Computational Chemistry Consortium (C3) is dedicated to leading the advancement of combustion and emissions modeling. The C3 cluster combines the expertise of different groups involved in combustion research aiming to refine existing chemistry models and to develop more efficient tools for the generation of surrogate and multi-fuel mechanisms, and suitable mechanisms for CFD applications. In addition to the development of more accurate kinetic models for different components of interest in real fuel surrogates and for pollutants formation (NOx, PAH, soot), the core activity of C3 is to develop a tool capable of merging high-fidelity kinetics from different partners, resulting in a high-fidelity model for a specific application. A core mechanism forms the basis of a gasoline surrogate model containing larger components including n-heptane, iso-octane, n-dodecane, toluene and other larger hydrocarbons.

Gasoline particulate filters (GPF) recently entered the market, and are already regarded a state-of-the-art solution for gasoline exhaust aftertreatment systems to enable EU6d-TEMP fulfilment and beyond. Especially for coated GPF applications, the prognosis of the emission conversion performance over lifetime poses an ambitious challenge, which significantly influences future catalyst diagnosis calibrations. The paper presents key-findings for the different GPF application variants. In the first part, experimental GPF ash loading results are presented. Ash accumulates as thin wall layers and short plugs, but does not penetrate into the wall. However, it suppresses deep bed filtration of soot, initially decreasing the soot-loaded backpressure. For the emission calibration, the non-linear backpressure development complicates the soot load monitoring, eventually leading to compromises between high safety against soot overloading and a low number of active regenerations.

To achieve global market and brand specific drivability characteristics as unique selling proposition for the increasing number of passenger car derivatives, an objectified evaluation approach for the drivability capabilities of the various cars is required. Thereto, it is necessary to evaluate the influence of different engine concepts in various complex and interlinked powertrain topologies during engine load change maneuvers based on physical criteria. Such an objectification approach enables frontloading of drivability related engineering tasks by the execution of drivability development and calibration work within vehicle subcomponent-specific closed-loop real-time co-simulation environments in early phases of a vehicle development program. So far, drivability functionalities could be developed and calibrated only towards the end of a vehicle development program, when test vehicles with a sufficient level of product maturity became available.

This article discusses highly flexible and accurate physics-based mean value modeling (MVM) for internal combustion engines and its wide applicability towards virtual vehicle calibration. The requirement to fulfill the challenging Real Driving Emissions (RDE) standards has significantly increased the demand for precise engine models, especially models regarding pollutant emissions and fuel economy. This has led to a large increase in effort required for precise engine modeling and robust model calibration. Two best-practice engine modeling approaches will be introduced here to satisfy these requirements. These are the exclusive MVM approach, and a combination of MVM and a Design of Experiments (DOE) model for heterogeneous multi-domain engine systems.

To reduce hydrocarbon and particle emissions as well as irregular combustion phenomena, the identification and quantification of possible oil sources in the combustion chamber of the direct injection gasoline engine are of main interest. The aim of this research activity is to fundamentally investigate the formation of locally increased lubricating oil concentration in the combustion chamber. For this purpose, the oil sources are considered separately from each other and divided into two groups - piston/compression ring and lubricating film on the liner. The associated oil emissions and their influence on the engine combustion process are the core of the investigations.