Skip to main content
main-content
Top

About this book

The majority of 0D/1D knock models available today are known for their poor accuracy and the great effort needed for their calibration. Alexander Fandakov presents a novel, extensively validated phenomenological knock model for the development of future engine concepts within a 0D/1D simulation environment that has one engine-specific calibration parameter. Benchmarks against the models commonly used in the automotive industry reveal the huge gain in knock boundary prediction accuracy achieved with the approach proposed in this work. Thus, the new knock model contributes substantially to the efficient design of spark ignition engines employing technologies such as full-load exhaust gas recirculation, water injection, variable compression ratio or lean combustion.

About the Author
Alexander Fandakov holds a PhD in automotive powertrain engineering from the Institute of Internal Combustion Engines and Automotive Engineering (IVK) at the University of Stuttgart, Germany. Currently, he is working as an advanced powertrain development engineer in the automotive industry.

Table of Contents

Frontmatter

Chapter 1. Introduction

The ongoing trend of enforcing lower fleet consumption and tightening emission standards, combined with ever-increasing customer demands and severe competition on the global market is the main driving force for further improvement of the internal combustion engine regarding its efficiency and emissions. Additional challenges are posed by the ongoing powertrain hybridization, as it imposes a significant improvement of the efficiency over the entire engine map [103]. Hence, new concepts are needed to guarantee the clean and energy-saving engine operation in a very wide range of operating conditions, especially at high loads, where knock typically occurs [122].
Alexander Fandakov

Chapter 2. Fundamentals and State of the Art

Nowadays 0D/1D simulations are being widely used in the engine development process. Thanks to the high prediction quality of the models and the low computational times, this is a powerful tool used to reduce development costs by partially eliminating the need for cost-intensive test bench investigations. This chapter presents the fundamentals of the 0D/1D engine simulation and reviews the existing knock modeling approaches.
Alexander Fandakov

Chapter 3. Experimental Investigations and Thermodynamic Analysis

For the development of a new 0D/1D knock model, the occurrence of knock has been extensively investigated on an engine test bench. To this end, a homogeneously operated direct injection spark ignition single-cylinder research engine featuring external boosting, low-pressure exhaust gas recirculation and a tumble generation device was used. This chapter briefly presents the experimental setup and describes the measurement data processing.
Alexander Fandakov

Chapter 4. Unburnt Mixture Auto-Ignition Prediction

As it is commonly assumed that knock occurs because of local auto-ignition(s) in the unburnt mixture, Sections 2.1.2 and 2.3.2 the accurate prediction of auto-ignition is the key to developing a fully predictive knock model for the 0D/1D engine simulation. This chapter aims at better understanding the chemical processes resulting in auto-ignition and developing an appropriate approach for predicting the occurrence of this phenomenon.
Alexander Fandakov

Chapter 5. Knock Occurrence Criterion

The transition to knock in SI engines does not only depend on the temperature and pressure in the end-gas [88]. The correct prediction of local auto-ignition is not sufficient for the reliable calculation of the knock boundary, as the occurrence of this phenomenon does not necessarily result in knock [88] [89]. To this end, an additional criterion for occurrence of knock resulting from the predicted auto-ignition is needed, which defines the main task in this chapter.
Alexander Fandakov

Chapter 6. Knock Model Overview

The present chapter gives an overview of the developed 0D/1D knock prediction approach as well as some remarks on the model use. To begin with, the possible sources for the model inputs needed for the estimation of the knock boundary are reviewed and the prediction quality they result in is evaluated. Additionally, the model versatility is investigated to ensure that the developed approach can be used in combination with various simulation models for the estimation of parameters such as the wall heat transfer.
Alexander Fandakov

Chapter 7. Knock Model Validation

In this chapter, the newly developed 0D/1D knock model is applied to available measurement data to assess its prediction accuracy and the results are discussed. The performance of the newly developed knock model has been evaluated extensively at various operating conditions on three different engines. The validation process involved the comparison of the measured MFB50-points at the knock boundary with the ones predicted by the knock model in the course of 0D/1D engine simulations, as the combustion center is decisive for the engine efficiency [75] [122].
Alexander Fandakov

Chapter 8. Conclusions and Outlook

Overall, a lot of effort has been put into the development of a reliable 0D/1D knock model in the past decades. The huge number of researches that have worked and are still working on this topic as well as the regularly communicated poor knock prediction accuracy suggested that major changes in the simplified approaches for modeling the real chemical and physical processes are needed in order to improve the knock prediction performance. This required the better understanding of the processes that lead to knock in real engines.
Alexander Fandakov

Backmatter

Additional information

Premium Partner

image credits