Search Results

An improved model-based two-degree of freedom control system for the intake manifold pressure in passenger car diesel engines is described in this paper. The aim of this control system is to track the air charge setpoint rapidly and precisely. To achieve this, an inverse model of the intake manifold dynamics is included in the feedforward control path. The system parameters which are necessary to calculate the inverse model are setpoints from other control loops in the gas system. These generated setpoint values allow for decoupling of the individual control loops in the gas system as far as possible. The parallel linear feedback controller is designed to further improve the accuracy of the control system. The calculated feedforward control signal and the feedback control variable additively generate the effective opened area of the intake throttle valve.

The efforts of the ISO “Test Variability Task Force” have been aimed at improving the understanding and at reducing brake dynamometer test variability during performance testing. In addition, dynamometer test results have been compared and correlated to vehicle testing. Even though there is already a vast amount of anecdotal evidence confirming the fact that different procedures generate different friction coefficients on the same brake corner, the availability of supporting data to the industry has been elusive up to this point. To overcome this issue, this paper focuses on assessing friction levels, friction coefficient sensitivity, and repeatability under ECE, GB, ISO, JASO, and SAE laboratory friction evaluation tests.

Two substantially different methods have become popular in building fast computing catalyst models: physico-chemical approaches focusing on dimensionality reduction and machine learning approaches. Data driven models are known to be very fast computing and to achieve high accuracy but they can lack of extrapolation capability. Physico-chemical models are usually slower and less accurate but superior regarding robustness. The robustness can even be reinforced by implementing an extended Kalman filter, which enables the model to adapt its states based on actual sensor values, even if the sensors are drifting. The present study proposes a combination of both approaches into one hybrid model, keeping the robustness of the physico-chemical model in edge cases while also achieving the accuracy of the data based model in well-known regimes. The output of the hybrid model is controlled by an autoencoder, utilizing methods well known from the field of anomaly detection.

The objective of this investigation was to evaluate the effects of a variable intake and exhaust valve timing in terms of opening, closing, opening duration, lift curve and number of active valves per pair on a four cylinder direct-injecting SI engine for the catalyst heating idling phase at the beginning of an NEDC emission test procedure. The first step evaluated the engine behavior at a reference point of operation. Its parameters in valve timing were adjusted to match the valve timing of the base production engine. The second step investigated the effects of an earlier exhaust valve opening while the exhaust valve closing time was kept and the exhaust valve opening duration was extended. The third step was to answer the question for the optimum number of exhaust valves in order to minimize the wall heat losses inside the cylinder head. The optimum 3V exhaust valve timing has been defined as the basis for exhaust valve timing for steps four and five.

A trip-based energy management strategy for electric vehicles (EVs) is proposed. It can use deterministic routing information obtained from, nowadays, available navigation systems and determines stochastic descriptions of process uncertainties such as stop events as unpredictable disturbances. A dynamic programming algorithm is used to calculate the optimal control trajectories required to reach the target destination safely and to suggest the driver an optimal driving style to maximize the battery range. The algorithm is implemented on a rapid prototyping platform using MATLAB/Simulink. Simulations and experimental results obtained from an EV prototype car are presented.

This paper proposes and evaluates improvements to a crash simulation of a fuel delivery module in a fuel tank. The simulations were performed in ANSYS/LS-DYNA. Deviations between the original simulation and test data were studied and reasons for the deviations hypothesized. These reasons stemmed from some of the simplifying assumptions of the model. Improvements consisted of incorporating plasticity and strain rate effects into the material models. Performance criteria were also directly incorporated into the material models such that non-performing portions of the model could be deactivated during the simulation. Finally, solid-fluid interactions were added into the simulation to include the momentum transfer from fuel to the fuel delivery module. It was previously thought that effects of a crash would be most severe on the module when the fuel tank was empty and the module was full with fuel.

Networking of active and passive safety is the fundamental basis for comprehensive vehicle safety. Situation-relevant information relating to driver reactions, vehicle behavior and traffic environment are fed into a crash probability calculator, which continually assesses the current crash risk and intervenes when necessary with appropriate measures to avoid a crash and reduce potential injuries. This provides effective protection not only for vehicle occupants but also for other, vulnerable road users. As this functionality up till now only relates to the vehicle itself, the next logical step is enhancement leading to the ultimate goal in safety performance, telematics. The integration of this embedded, in-vehicle wireless communication system allows Car-to-Car (C2C) and Car-to-Infrastructure (C2I) functionality for, e.g. hazard warning. This is an integral element of the cascaded ContiGuard® protection measures.

The AUTomotive Open System ARchitecture (AUTOSAR) Development Partnership has published early 2008 the specifications Release 3.0 [1], with a prime focus on the overall architecture, basic software, run time environment, communication stacks and methodology. Heavy developments have taken place in the OEM and supplier community to deliver AUTOSAR loaded cars on the streets starting 2008 [2]. The 2008 achievements have been: Improving the specifications in order to secure the exploitation for body, chassis and powertrain applications Adding major features: safety related functionalities, OBD II and Telematics application interfaces.

There are numerous methods for accelerated ash loading of particulate traps known from literature. However, it is largely unknown if a combination of these methods is possible and which one generates the most similar ash compared to ash from real particulate filters. Since the influencing variables on the ash formation are not yet fully understood, ashing processes are carried out under carefully controlled laboratory conditions on an engine test bench. The first ashing takes place with low sulfated ash phosphorus and sulfur oil without any methods to increase the quantity of produced ash. The obtained ash is used as a reference and is compared hereinafter with the process examined. Four methods to increase the ash production ratio are investigated. The first one is an increase of the ash content of the lubrication oil through an increase of the additives in the oil. The second one is the additional generation of ash with a burner system where oil is injected into the flame.

In the research project between the Institute of Automotive Engineering (FZD) of the Technische Universität Darmstadt (TUD) and Continental Teves AG & Co. oHG a new modeling concept has been developed. With the aim to enhance the current development process, the brake caliper is modeled based on coupled rigid bodies integrated into a nonlinear system model. Using an explicit interface definition, the number of degrees of freedom is minimized and the calculation of caliper performance is possible over a wide range of parameters. Compared to models based on the Finite Element Method (FEM), fully parameterized geometry from CAD is not necessary, thus the caliper can be optimized for a variation of its geometrical and physical parameters. With this modeling approach, typical performance criteria such as caliper fluid displacement, hysteresis, uneven pad wear and residual torque can be calculated in a virtual bench test.

In this paper optimal control problems for hybrid powertrain vehicles with different drive-modes are considered and solved using numerical techniques. This leads to the formulation of hybrid optimal control problems. The aim is to find optimal controls and optimal switchings between the drive-modes to minimize a cost function resembling fuel consumption. The problem is nonlinear and subject to constraints concerning both controls and state. The techniques include indirect methods as well as direct optimization methods. Efficiency and accuracy are evaluated for all methods using simulation studies. An experimental test on a near mass-production vehicle confirms the usability of the direct optimization approach.

The quality of the powertrain design has a significant impact on the fuel consumption and emissions of hybrid vehicles. Lack of experience with these relatively new technologies, the enormous variety of hybrid powertrain configurations, and the multitude of components make this area an ideal application for computer-based modeling and optimizations. Global optimization techniques have the advantage to explore systematically the design space to find the optimal configuration space. In this paper, a systematic procedure for an optimal design of hybrid powertrain configurations using an evolutionary algorithm is proposed. It will be shown that the design steps for parallel and power-split configurations are quite similar. This results in a computing approach with high synergy effects and the ability to exchange components seamless to compare different ‘virtual’ configurations.

The US American government introduced a law to mandatorily equip passenger vehicles with rear view cameras. Furthermore, US NCAP presented a test for passenger vehicles to brake on pedestrians while back up. These two circumstances lead to main motivation of the development of the Comfort Backup Assist (CBUA). Nevertheless, more and more passenger cars in general are being equipped with rear view cameras. Rear view system (RVS) allows to deliver a rear-view camera system including a braking functionality which is intended to make the driving mission safer and reduce the number of accidents in parking driving situations. RVS also focus on vehicle safety by reducing accidents while taking reversing/parking scenarios and to provides slow de-acceleration of the vehicle gradually to avoid jerk and increase the ride comfort.

Driver assistance features have been introduced to the market focusing on basic, independent functional scenarios. The trend is showing that these kinds of products are facing more and more complex scenarios and we are transitioning from single independent functions to a strongly networked system. Some of the drivers for future autonomous vehicles are 360° monitoring by active safety technology and V2X (vehicle to vehicle or vehicle to infrastructure) communication. In the past vehicles were strictly operated by the driver. Advanced driver assistance products added so called feedback features like lane departure warning, forward collision warning, and blind spot monitoring. First steps towards semi-autonomous driving started with the development of active support functions like adaptive cruise control or lane keeping support. Collision mitigation with various authority levels is the next milestone towards automation followed by other, even more advanced, features.

Compressed natural gas (CNG) is an attractive, alternative fuel for spark-ignited (SI), internal combustion (IC) engines due to its high octane rating, and low energy-specific CO2 emissions compared with gasoline. Directly-injected (DI) CNG in SI engines has the potential to dramatically decrease vehicles’ carbon emissions; however, optimization of DI CNG fueling systems requires a thorough understanding of the behavior of CNG jets in an engine environment. This paper therefore presents an experimental and modeling study of DI gaseous jets, using methane as a surrogate for CNG. Experiments are conducted in a non-reacting, constant volume chamber (CVC) using prototype injector hardware at conditions relevant to modern DI engines. The schlieren imaging technique is employed to investigate how the extent of methane jets is impacted by changing thermodynamic conditions in the fuel rail and chamber.

The contribution presents a new structurally optimized 7-speed hybrid DCT for transverse applications. Advantageously the seven forward speeds can be provided with only two shafts. The space of the usually third shaft can be used for the electric motor or the mechanical reverse speed. This modular integration can be realized by using mostly identical parts for the components of the gear set and the transmission housing. The direct connection of the electric motor to the differential helps to avoid additional gear set loads and efficiency losses. Such modular transmissions will help in future to reduce development and production costs.

Accuracy of clutch torque model which converts target torque to target stroke is essential to control the dry clutch system. Continuous Adaptation algorithm requires micro slip control during in-gear driving. Clutch judder during micro slip control can cause detrimental effect on the output of controller as slip speed is calculated by deviation of engine speed and clutch speed. Conventional approach to avoid clutch judder is using low pass filter to the input of controller which is slip speed. But this affect to the overall response time of slip controller. In this paper, signal processing algorithm is design and tested for the clutch speed(Input shaft speed). With low pass filter in clutch speed, clutch judder signal is decreased but overall time delay creates static error during acceleration. Several phase advance algorithm is designed to overcome the static error during acceleration without disadvantage of decreasing clutch judder signal.

In this paper, a hybrid optimal control problem (HOCP) for the optimal heating of the three-way catalytic converter is solved. We propose a model for a hybrid vehicle that beneath State of Charge and fuel consumption includes thermal system states like engine cooling water temperature and catalytic converter temperature. Since models for noxious emissions with appropriate computational demand are not yet available for optimization purposes, an artificial state that resembles the emissions produced is introduced. A hybrid optimal control problem is then formulated for the beginning of the FTP-75 drive cycle whose target is to minimize the energy loss during the catalytic converter and engine cooling water heating phase. The corresponding input values to be optimized are continuous variables as ignition angle and cylinder charge as well as discrete decisions such as different injection schemes. As additional constraint, an upper limit is imposed on the artificial emissions state.

Global automotive fuel economy and emissions pressures mean that 48 V hybridisation will become a significant presence in the passenger car market. The complexity of powertrain solutions is increasing in order to further improve fuel economy for hybrid vehicles and maintain robust emissions performance. However, this results in complex interactions between technologies which are difficult to identify through traditional development approaches, resulting in sub-optimal solutions for either vehicle attributes or cost. The results presented in this paper are from a simulation programme focussed on the optimisation of various advanced powertrain technologies on 48 V hybrid vehicle platforms. The technologies assessed include an electrically heated catalyst, an insulated turbocharger, an electric water pump and a thermal management module.

In this paper the power hop phenomenon is analyzed and important influencing factors are investigated. The results of driving tests on various road surfaces with different types of cars with longitudinal and transversal mounted engines as well as with front and rear wheel drive are presented. In order to understand and quantify the power hop effect the rotational speed of the individual wheels and the engine are measured. Additionally, the drive shaft torque, the engine movement in its bearings and the vertical deflection of the wheel with respect to the chassis are determined to get detailed knowledge about physical dependencies. It is shown that the rotational speed of the driven wheels is not a sufficient indicator to assess the occurrence of power hop by measurements. Alternatively, the measured longitudinal acceleration at the seat rail provides a good quantification.