Search Results

One of the most critical challenges currently facing the diesel engine industry is how to improve fuel economy under emission regulations. Improvement in fuel economy can be achieved by precisely controlling Air/Fuel ratio and by monitoring fuel consumption in real time. Accurate and repeatable measurements of fuel rate play a critical role in successfully controlling air/fuel ratio and in monitoring fuel consumption. Volumetric and gravimetric measurements are well-known methods for measuring fuel consumption of internal combustion engines. However, these methods are not suitable for obtaining fuel flow rate data used in real-time control/measurement. In this paper, neural networks are used to solve the problem concerning discontinuous data of fuel flow rate measured by using an AVL 733 s fuel meter. The continuous parts of discontinuous fuel flow rate are used to train and validate a neural network, which can then be used to predict the discontinuous parts of the fuel flow rate.

Traditionally, university research in engine technology has been focused on fundamental engine phenomena. Increasingly however, research topics are developing in the form of systems issues. Examples include air and exhaust gas recirculation (EGR) management, after-treatment systems, engine cooling, hybrid systems and energy recovery. This trend leads to the need for engine research to be conducted using currently available products and components that are re-configured or incrementally improved to support a particular research investigation. A production engine will include an electronic control unit (ECU) that must be understood and utilised or simply removed and circumvented. In general the intellectual property (IP) limitations places on ECUs by their suppliers mean that they cannot be used. The supplier of the ECU is usually unable to reveal any detail of the implementation. As a consequence any research using production hardware is seriously disadvantaged from the beginning.

Variable valve actuation in heavy duty diesel engines is not well documented, because of diesel engine feature, such as, unthrottled air handling, which gives little room to improve pumping loss; a very high compression ratio, which makes the clearance between the piston and valve small at the top dead center. In order to avoid strike the piston while maximizing the valve movement scope, different strategies are adopted in this paper: (1) While exhaust valve closing is fixed, exhaust valve opening is changed; (2) While exhaust valve closing is fixed, late exhaust valve opening: (3) While inlet valve opening is fixed, inlet valve closing is changed; (4) Delayed Inlet valve and exhaust valve openings and closings; (5) Changing exhaust valve timing; (6) changing inlet valve timing; (7) Changing both inlet and exhaust timing, will be used.

This work describes the application of Non-Linear Autoregressive Models with Exogenous Inputs (NLARX) in order to predict the NOx emissions of heavy-duty diesel engines. Two experiments are presented: 1.) a Non-Road-Transient-Cycle (NRTC) 2.) a composition of different engine operation modes and different engine calibrations. Data sets are pre-processed by normalization and re-arranged into training and validation sets. The chosen model is taken from the MATLAB Neural Network Toolbox using the algorithms provided. It is teacher forced trained and then validated. Training results show recognizable performance. However, the validation shows the potential of the chosen method.

Variable compression ratio in conjunction with a control system is an effective way to improve performance and reduce emissions in a diesel engine. There are various methods that may be employed that include geometry changes and varying valve timing to change the effective compression ratio. In this paper, a simulation study is presented that is based on a modern, multi-cylinder, fixed compression ratio diesel engine equipped with exhaust gas recirculation (EGR) and a variable geometry turbocharger (VGT). The engine is represented using the GT-Power code, and includes a predictive combustion model. The aim of the investigation is to identify the impact of variable compression ratio on fuel economy and emission reduction and whether realistic optimal conditions exist. This paper describes how a formal design of experiments procedure is used to define the simulation conditions. Cost functions are defined with different weights for fuel consumption, NOx and soot emissions.

More and more stringent emission regulations require advanced control technologies for combustion engines. This goes along with increased monitoring requirements of engine behaviour. In case of emissions behaviour and fuel consumption the actual combustion efficiency is of highest interest. A key parameter of combustion conditions is the in-cylinder pressure during engine cycle. The measurement and detection is difficult and cost intensive. Hence, modelling of in-cylinder conditions is a promising approach for finding optimum control behaviour. However, on-line controller design requires real-time scenarios which are difficult to model and current modelling approaches are either time consuming or inaccurate. This paper presents a new approach of in-cylinder condition prediction. Rather than reconstructing in-cylinder pressure signals from vibration transferred signals through cylinder heads or rods this approach predicts the conditions.

For diesel engines, fuel path control plays a key role in achieving optimal emissions and fuel economy performance. There are several fuel path parameters that strongly affect the engine performance by changing the combustion process, by modifying for example, start of injection and fuel rail pressure. This is a multi-input multi-output problem. Linear Model Predictive Control (MPC) is a good approach for such a system with optimal solution. However, fuel path has fast dynamics. On-line optimisation MPC is not the good choice to cope with such fast dynamics. Explicit MPC uses off-line optimisation, therefore, it can be used to control the system with fast dynamics.