Refine
Document Type
Has Fulltext
- no (6)
Reviewed
Keywords
Institute
Is part of the Bibliography
- yes (6)
A novel one-dimensional multiphase and multicomponent spray model - hereafter referred to as the Kattke-Weigand model - has been developed to predict the penetration length of both vapor and liquid gasoline sprays under flash-boiling conditions, such as superheated injections. Its formulation is based on mass and momentum equations for unsteady jets and is therefore capable of capturing dynamic effects. Experiments were conducted in a constant volume chamber using various ambient and fuel temperature conditions and a six-hole GDI injector with a separated jet. Macroscopic spray parameters were extracted from the measurements to verify the model's ability to predict both liquid and vapor penetration length and the corresponding spray angles. Apart from the separated jet of the injector used, the other five jets interact strongly with each other under flash boiling conditions, resulting in spray collapse, and thus affecting spray characteristics. The prediction of collapse is very sensitive to calculations of vaporization and air entrainment. Since these submodels cannot be validated directly, a calibration method as developed, that is based on a three-dimensional reconstruction of all fuel sprays of the injector used. For this purpose, all optical measurements performed in the constant volume chamber are utilized. As a result, a three-dimensional representation of the spray collapse can be calculated from the combination of the 3D spray reconstruction and the entrainment and vaporization submodels. The validation of the collapse leads indirectly to the calibration of the entrainment and vaporization submodels in the Kattke Weigand model. Latter is applied to gain a deeper understanding of the interaction between spray collapse and both liquid and vapor phase penetration.
A novel one-dimensional multiphase and multicomponent spray model - hereafter referred to as the Kattke-Weigand model - has been developed to predict the penetration length of both vapor and liquid gasoline sprays under flash-boiling conditions, such as superheated injections. Its formulation is based on mass and momentum equations for unsteady jets and is therefore capable of capturing dynamic effects. Experiments were conducted in a constant volume chamber using various ambient and fuel temperature conditions and a six-hole GDI injector with a separated jet. Macroscopic spray parameters were extracted from the measurements to verify the model's ability to predict both liquid and vapor penetration length and the corresponding spray angles. Apart from the separated jet of the injector used, the other five jets interact strongly with each other under flash boiling conditions, resulting in spray collapse, and thus affecting spray characteristics. The prediction of collapse is very sensitive to calculations of vaporization and air entrainment. Since these submodels cannot be validated directly, a calibration method was developed, that is based on a three- dimensional reconstruction of all fuel sprays of the injector used. For this purpose, all optical measurements performed in the constant volume chamber are utilized. As a result, a three-dimensional representation of the spray collapse can be calculated from the combination of the 3D spray reconstruction and the entrainment and vaporization submodels. The validation of the collapse leads indirectly to the calibration of the entrainment and vaporization submodels in the Kattke Weigand model. Latter is applied to gain a deeper understanding of the interaction between spray collapse and both liquid and vapor phase penetration.
Harmful NOx emissions generated during rapid changes in engine’s speed and load (such as acceleration or deceleration) are strongly linked to deviations in air-to-fuel ratio (λ) from its optimal setpoint. These deviations arise under transient conditions due to several factors, including sensor resolution limitations, time delays between actuators and sensors, and inaccurate air mass estimation within the engine management system.
While fuel mass control has significantly improved over the past three decades - achieving superior performance in terms of both dynamics and accuracy through the use of multiple injections per engine cycle - the primary challenge remains the precise control of air mass, particularly under transient conditions. Nonlinear wave dynamics and the inertia of the engine’s air path make traditional solutions, such as slow and inefficient actuators like the butterfly valve, inadequate.
This paper introduces a novel methodology to address these limitations by developing a device that is an order of magnitude faster than a butterfly valve while minimizing flow losses. The design philosophy is outlined, and the flow bench setup used to evaluate the performance of an instrumented prototype is presented. High-frequency data collected during testing are analyzed, and the extracted flow structure and turbulence parameters are discussed.
Finally, the dynamic performance of the newly developed device is incorporated into a transient gas exchange engine model, where its capability to control λ on a cycle-resolved basis is evaluated against a conventional throttle body. Numerous studies have proven that NOx spikes during engine’s accelerations and decelerations are well correlated to λ deviations from its setpoint. The assessment of the emissions reduction strategy under transient conditions is based on the device's precision in managing λ during each engine cycle, as derived from the transient model output.
Over the last decade, the prevailing demand for renewable energy sources initiated the electrification of the transportation sector, with the development of H2 engines and fuel cell powertrains increasingly gaining momentum alongside battery electric vehicles. One of the essential tasks to optimize the overall efficiency of these drives is to precisely, dynamically and frictionless meter the reacting gases into the conversion chambers. In this paper, we propose a novel dosing system, focusing on gases (air), that is based on the principle of a variable Venturi nozzle, which incorporates a device for rapid variation of the effective flow cross-sectional area (EA) with negligible pressure losses. Each EA corresponds to a demanded load or mass flow and thus enables a dynamic control of the stoichiometry (l) of the machine. Our patented technology mainly consists of two surfaces perpendicular to the flow direction where one is fixed in space and the other one is moving at high frequency (> 20 Hz) towards or against the flow direction. Firstly, the underlying design philosophy and the details of the first prototype are presented, followed by an extensive description of the test bench setup, including data acquisition system. This is succeeded by a thorough analysis of the high-frequency data to extract flow characteristics and turbulence parameters. Subsequently, the transient performance of the Venturi system is experimentally compared with that of a throttle body. Finally, the capability of the Venturi system to precisely control the cycle-resolved air-fuel ratio of a hydrogen engine is demonstrated through transient simulation results based on a class B car operating under the FTP-75 driving cycle.