Xin Q. Diesel Engine System Design

Подождите немного. Документ загружается.

96 Diesel engine system design

area. Another example is an excessively high exhaust manifold gas temperature

caused by cooler failure or low air–fuel ratio. Generally, the need of root cause

analysis should be minimized as much as possible through successful up-

front specication design with system optimization. The system performance

specication is usually the most important of the four system specications

in engine design (performance, durability, packaging, and cost).

During the process of generating engine system design specications, often

off-the-shelf solutions or design need to be considered in order to simplify

the product design. The off-the-shelf design usually has been fully tested

on a specic engine. Its cost and manufacturing method are also known.

Engine system design needs to verify the selected off-the-shelf solution prior

to testing validation.

1.7.3 System durability

The durability design specication is expressed in terms of both ‘stress’ and

‘strength’. The stress refers to any general loading, usually coming from the

performance specication (e.g., peak cylinder pressure, cylinder heat ux,

exhaust manifold temperature, compressor outlet temperature). The strength

refers to the desirable structural design parameters representing the capability

of the system or component. The strength is also used by the performance area

as a design constraint for iteratively rening the performance specication.

Examples of preliminary engine structural calculations were provided by

Makartchouk (2002) and Delprete et al. (2009).

1.7.4 System packaging

System packaging addresses the issues with weight, size, shape, component

location, and clearance between the components. Good engine packaging

requires low weight, compactness, suitable shape t in the engine compartment,

reasonable relative locations and clearances between the components without

functional or geometrical interferences. Engine packaging design is handled

mainly by three-dimensional solid modeling.

Engine weight can be estimated by using the volumes of the components.

The representative dimensions of major components can be classied into

two categories: (1) the dimensions determined by the basic design parameters

of the engine such as cylinder bore diameter and engine stroke; and (2) the

dimensions determined by the durability requirements that are related to engine

operating speed and cylinder pressure loading. Weight, size, and cost are

especially important for NVH design when seeking an acoustically-effective

solution to package. Schuchardt et al. (1993) introduced the concepts of

packaging quantity and quality and the relationship between packaging and

performance in their study on engine intake noise control.

Diesel-Xin-01.indd 96 5/5/11 11:41:38 AM



97The analytical design process and diesel engine system design

Engine and component size and the overall shape are strongly affected

by the engine conguration (i.e., number of cylinders, inline, Vee, or

opposed arrangement), bore, stroke, connecting rod length, valvetrain type

(pushrod or overhead cam), turbocharger type (single-stage, two-stage, or

twin-parallel), EGR cooler size, and FEAD. Component size is also related

to performance. This is especially true for diesel aftertreatment devices. The

device size is usually limited by the available spacing of the under-oor

for a given vehicle. A smaller aftertreatment component has higher space

velocity and may have lower operating efciency. For example, there is a

trade-off between LNT size and fuel consumption penalty. A smaller size

gives higher fuel consumption penalty.

The design of relative locations and clearances between the components

deserves great attention. The performance and durability characteristics of

the engine components are often subject to the harsh thermo-mechanical

boundary conditions. They are affected by the heat transfer and vibration

of the neighboring components. A primary focus of system packaging work

is to optimize the relative positions of all the subsystems or components in

order to minimize the negative impact of all thermo-mechanical effects as

a whole.

Lastly, it should be noted that in modern diesel engine system design,

design for manufacturability and design for serviceability are two important

trends in packaging design. They represent the processes to optimize the

relationships among design function, manufacturability, ease of assembly,

and ease of maintenance.

1.7.5 System cost

Engine system cost can be estimated by using bill of materials and operating

cost. The attributes of performance, durability and packaging all directly

affect cost. The engine system cost specication plans the capital cost and

operating cost of the engine technologies adopted. It also coordinates between

the subsystems on the maximum allowable cost for each subsystem in order

to control the total engine cost.

1.8 Work processes and organization of diesel

engine system design

1.8.1 Characteristics and principles of diesel engine

system design

Academic background and characteristics of diesel engine system design

A diesel engine is a mechanical system. According to the denition of

‘mechanical system’ given by the National Science Foundation (Panel Steering

Diesel-Xin-01.indd 97 5/5/11 11:41:38 AM



98 Diesel engine system design

Committee, 1984), a mechanical system can be dened as an interconnection of

mechanical and/or electromechanical components, coordinated and controlled

by computational and informational networks (and often humans), which

accomplishes dynamic tasks involving mechanical forces and motions and

energy ows. The eld of mechanical systems can be further classied into

the following four major academic disciplines: design methodology and

interactive graphics; dynamic systems and control; machine dynamics; and

tribology. The discipline of design methodology deals with computer-aided

engineering, optimization, and other generic techniques. The discipline of

dynamics systems and control deals with the dynamic behavior and control

of the interacting subsystems or components, including their energy ows,

motions, and forces. Machine dynamics involves kinematics, dynamics,

solid mechanics, acoustics, and nite element methods, and are generally

related to reliability/durability. Tribology is the science and technology of

lubrication, friction, and wear.

Diesel engine system design is a multi-disciplinary application eld

which possesses the features of the above academic disciplines. First of all,

it requires an optimization approach supported by systems engineering and

robust engineering. Secondly, it requires an approach of system dynamics to

handle both transient (dynamic) performance/control and steady-state (static)

performance. Thirdly, it requires dynamic motion analysis for the critical

systems such as vehicle longitudinal dynamics, piston-assembly dynamics

and valvetrain dynamics. Lastly, it is related to friction analysis of the engine

components since friction directly affects engine efciency.

A diesel engine system is also a combination of hardware and software

control. It requires proper hardware sizing and matching to achieve target

performance, and it also requires electronic software control to realize the

performance during fast-changing and highly-nonlinear transient events. A

correct system design can be achieved only when high-quality analytical

tools or simulation models are available.

Moreover, a diesel engine is a thermodynamic energy system which

produces power by a working uid medium. The system performance design

specications address mainly the gas pressures, temperatures, and ow rates

inside the engine. Advanced energy system analysis and design require using

both the thermodynamic rst law and second law. The rst law provides

the energy distribution inside the system, while the second law reveals the

irreversibility, availability and the maximum potential of each component

and process of the system.

Diesel engine system design and EPSI

Diesel engine system design includes four major branches: performance,

durability, packaging, and cost. Performance is the leading one and carries

Diesel-Xin-01.indd 98 5/5/11 11:41:38 AM



99The analytical design process and diesel engine system design

the majority of the work in the entire system design. The complete name of

the performance branch in diesel engine system design is engine performance

and system integration (EPSI). The function of EPSI was illustrated earlier

in Fig. 1.27. EPSI analyzes and integrates the performance of various engine

subsystems or technologies, and properly matches them with optimization

approaches. It conducts large-scale sophisticated computation in the areas

related to thermal conditions, uids, dynamics, and controls to derive precise

system design solutions by using analytical tools such as engine cycle simulation

software or analytical models and advanced data processing techniques.

System design is very important for the integration of simultaneous

engineering processes for diesel engine product design, ranging from high-

level product strategy planning to detailed production design. The missions

of diesel engine system design include:

∑ providing engine system design and analysis;

∑ developing system analysis methodologies and simulation techniques;

and

∑ developing core engine technologies with the viewpoint of system

integration.

Engine system design covers a wide range of technical specialties including

vehicle–engine–aftertreatment integration, thermodynamic cycle performance,

air system design and turbo matching, powertrain dynamics and electronic

controls.

1.8.2 Theoretical foundation and tools for diesel engine

system design

There are several tools used in diesel engine system design. The rst tool is

engine cycle simulation. Complete cycle simulation for engine performance

emerged in the early 1960s when digital computers became available. The

computer codes were pioneered by Benson and Woods (1960) and Borman

(1964). The initial use of engine cycle simulation was simple and restricted

mainly to research groups. As a result, engine design decisions at that time

were primarily driven by testing rather than computation. Since the 1980s,

engine cycle simulation has gradually moved from the research groups into

the production design and development process to support design (Morel and

LaPointe, 1994). Today’s engine development process requires optimized

design with much shorter development time than ever for much more

sophisticated engines. The analytical design process inevitably demands

engine cycle simulation to become a part of the standard design tools used

for production engine system design.

The foundation of engine system design analysis is built on thermodynamic

cycle simulation. Engine cycle performance models are either zero-dimensional

Diesel-Xin-01.indd 99 5/5/11 11:41:38 AM



100 Diesel engine system design

(i.e., spatially homogeneous, based on ordinary differential equations) or

one-dimensional (based on partial differential equations of manifold wave

dynamics). The input of the model includes engine geometry, subsystem

characteristics and engine calibration parameters such as fuel injection

timing and EGR valve opening. There are two types of output: the crank-

angle based instantaneous values (e.g., gas pressure, temperature, and ow

rate); and the cycle-average macro system performance parameters (e.g.,

engine torque, air–fuel ratio, and coolant heat rejection). Both steady-state

and transient performance can be computed at various ambient conditions.

There are several commercial software packages of cycle simulation, such as

Gamma Technologies’ GT-POWER (Morel et al., 1999), Ricardo’s WAVE

and AVL’s BOOST. Figure 1.33 shows an example of the engine cycle

simulation model. The key issues in the modeling for engine system design

are as follows:

∑ intake restriction (characterized by pressure drop vs. intake ow rate)

∑ exhaust restriction (characterized by pressure drop vs. exhaust ow

rate)

∑ engine intake and exhaust valves (characterized by instantaneous effective

valve ow area)

∑ intake and exhaust manifolds (characterized by volume, heat transfer,

and pressure drop frictional losses)

∑ cylinder (characterized by volumetric efciency, mechanical friction,

and base engine heat rejection)

∑ coolers (characterized by ow restriction and thermal effectiveness)

∑ EGR valve and intake throttle valve (characterized by ow restriction

through an orice)

∑ turbine (characterized by effective area and efciency)

∑ compressor (characterized by ow range and efciency).

More discussions on engine cycle simulation are provided in Chapter 4.

The second tool in engine system design is vehicle and powertrain

modeling. The model can be used to calculate the forces and torques in an

engine–drivetrain system and the transient motion of the vehicle. The main

analysis objectives include engine–transmission matching, hybrid powertrain

supervisory control, vehicle transient acceleration performance and driving

cycle fuel economy. The model accounts for road grade, rolling resistance,

aerodynamic drag, brake, powertrain inertia, clutch, torque converter and

transmission characteristics, drive axles, driver behavior, powertrain controls,

and engine performance characteristics (e.g., map-based mean-value model

or high-delity crank-angle-resolution model). The model usually can be run

in two different methods: forward (dynamic) or backward (kinematic). In the

forward method, the equations of motion for the drivetrain components and

the vehicle are numerically integrated in time to obtain the transient speeds

Diesel-Xin-01.indd 100 5/5/11 11:41:38 AM



Ambient

Pipes

Intake

port

Valves

Cylinder

Exhaust

port

EGR

cooler

Compressor

Cranktrain

EGR valve EGR-air mixing

Charge air cooler

Intake manifold

Exhaust

manifold

Turbine

Ambient

Aftertreatment

Cycle simulation model

Theory of the model

• Engine cycle thermodynamics

• Mass and energy conservation

• In-cylinder gas property changes

• Cylinder and pipe heat transfer

• Turbocharger principle

• Combustion and heat release rate

• 1-dimensional gas wave dynamics

• Acoustics of intake and exhaust

• Piston-assembly dynamics

• Controls linked to Matlab/Simulink

Input of the model

• Engine geometry

• Valve lift prole and port ow C

• Turbocharger maps and efciency

• Base engine heat rejection percentage

• Engine mechanical friction

• EGR cooler and CAC characteristics

• Fuel injection or combustion timing

Output of the model

• Cylinder average parameters (e.g., BSFC)

• Instantaneous gas pressure, temperature

and mass ﬂow rate inside the engine

1.33 Engine cycle simulation model (GT-POWER).

Diesel-Xin-01.indd 101 5/5/11 11:41:39 AM



102 Diesel engine system design

and torques in the system. One example is to predict 0–60 mph acceleration

with gear shifting events. In the backward method, the vehicle speed driving

prole is the input. The required speeds and torques of the powertrain

components to drive the vehicle are calculated based on the vehicle force

balance equation. One example is to calculate engine operating points for a

given driving cycle and the corresponding fuel consumption and emissions.

Typical vehicle simulation software packages include Gamma Technologies’

GT-DRIVE (Morel et al., 1999) and AVL’s Cruise. Other powertrain dynamics

models based on Simulink programming were introduced by Moskwa et

al. (1999) and Assanis et al. (2000). More discussions on engine–vehicle

matching are provided in Chapter 5.

The third tool is diesel aftertreatment modeling. The primary objectives

of aftertreatment analysis in diesel engine system design include the

following:

∑ simulating tailpipe emissions;

∑ conguration architecture selection;

∑ component sizing;

∑ precious metal loading;

∑ controlling DPF regeneration at all vehicle operating conditions for safe

and fuel efcient operation; and

∑ optimizing the interaction with other engine subsystems.

Over the past several years, considerable progress has been made to

model individual aftertreatment components such as DOC, DPF, LNT,

and SCR. For system design a more useful tool needs to be a system-level

aftertreatment model that is integrated with the engine model. Such integrated

diesel aftertreatment modeling was developed in Gamma Technologies’

GT-POWER by Tang et al. (2007) and Wahiduzzaman et al. (2007). Other

integrated models based on Simulink were introduced by Rutland et al. (2007)

and He (2007). The DOC model simulates the exhaust gas temperature raise

due to hydrocarbon oxidation caused by fuel dosing for DPF regeneration.

The DOC model includes chemical reaction kinetics for the oxidation of

CO, HC, and NO to CO

, H

O, and NO

. The DPF model can be either a

zero-dimensional (also called lumped-parameter) model or a one-dimensional

model. The one-dimensional model considers the axial changes of exhaust

gas dynamics and particulate matter oxidation. The input to the model is

exhaust gas ow from the turbine with heat losses through the tailpipe. The

DPF model simulates the deep-bed ltration, soot-cake ltration, particulate

matter deposition inside the lter and the resulting pressure drop. The DPF

regeneration process can be simulated by the model for thermal and catalytic

oxidations of the trapped soot. More discussions on engine–aftertreatment

integration are provided in Chapter 8.

The fourth tool is optimization and probabilistic analyses. The DoE

Diesel-Xin-01.indd 102 5/5/11 11:41:39 AM



103The analytical design process and diesel engine system design

method has been widely used to generate a matrix containing partial-factorial

combinations of input factors for efcient data gathering. Each case (or run)

in the DoE matrix can be simulated with engine cycle simulation software.

The output data are called responses. Mathematical models (polynomial

emulators) are built with response surface methodology (RSM (Myers and

Montgomery, 2002)) to link each response parameter with the factors. Then,

optimization is conducted by using the emulators to search the global optima

under certain constraints by minimizing or maximizing a functional target (e.g.,

minimizing BSFC). Probabilistic analysis can handle the system design task

of variability- or reliability-based optimization (Baker and Brunson, 2000).

The probability distributions of random input factors are generated with the

Monte Carlo simulation by producing thousands of cases. Each case is run

using a simulation model to produce response data. Then the thousands of

response data are processed with statistical software to obtain the probability

distribution of the response. Variability- or reliability-based design constraints

can be applied to conduct optimization in order to determine the optimum

mean and tolerance range of the control factors. The optimization and

probabilistic analysis methods will be introduced in detail in Chapter 3.

The trend in the vehicle industry is toward integrated engine–powertrain–

vehicle design and electronic controls. It is desirable to have an integrated

simulation environment containing all the engine system design tools for

seamless connections between the systems and easy data sharing. Gamma

Technologies has made impressive progress to integrate such tools within

the GT-SUITE environment (Ciesla et al., 2000; Silvestri et al., 2000; Morel

et al., 2003).

1.8.3 Technical areas of engine performance and system

integration

For the areas in diesel engine system design related to performance, two

categories of research are identied (Fig. 1.34). The rst category links the

system design to generic design methodologies, for example, how to apply

the principle of reliability engineering in diesel engine system design, or how

to formulate system dynamics models effectively to enhance the quality of

system design. The second category of research addresses the analysis and

modeling of each key subject area, and their interactions. Table 1.6 shows

the required skill sets in engine system design.

The functional areas in engine performance and system integration include

the following:

1. Technical eld 1: Engine system thermodynamic cycle performance

∑ Basic engine operation (power rating, torque curve, etc.)

∑ Basic engine design conguration (cylinder bore, stroke, etc.)

Diesel-Xin-01.indd 103 5/5/11 11:41:39 AM



104 Diesel engine system design

∑ Valvetrain performance (valve timing, cam, valvetrain dynamics)

∑ Variable valve timing

∑ Cylinder deactivation

∑ Exhaust restriction

∑ Combustion, fuel injection, and emissions modeling

∑ Cold start performance

∑ Turbocharging

∑ EGR system performance

∑ Engine heat rejection

∑ Cooling circuit performance

∑ Intake and exhaust manifold gas wave dynamics

∑ Engine friction

∑ Waste heat recovery

∑ Fuel economy improvement roadmap

∑ Engine brake performance

∑ Transient engine performance (load response, turbo lag, warm-up,

emissions cycles, etc.)

2. Technical eld 2: Vehicle and powertrain performance

∑ Vehicle acceleration, launch, gradeability, and drivability

Systems engineering

Robust engineering

Cost engineering

Vehicle performance

Valvetrain design

VVA performance

Engine friction

Waste heat recovery

Engine controls

Engine applications

Hybrid powertrain

Engine brake

Emissions modeling

Engine calibration

Optimization methods

System dynamics

System design process

and speciﬁcations

Engine cycle

performance

Thermodynamic ﬁrst

law and second law

Diesel aftertreatment

modeling & integration

Heat rejection and

cooling

Engine air system

(turbocharger & EGR)

Powertrain transient

performance

First category (methodology) Second category (subject areas)

1.34 Technical areas in diesel engine system design.

Diesel-Xin-01.indd 104 5/5/11 11:41:39 AM



105The analytical design process and diesel engine system design

∑ Drivetrain conguration and drive axle ratio

∑ Engine–transmission matching

∑ Vehicle driving cycle fuel economy

∑ Transient powertrain dynamics

∑ Hybrid powertrain performance and supervisory controls

3. Technical eld 3: Aftertreatment system integration

∑ Aftertreatment performance calibration and matching

∑ Aftertreatment emissions modeling

∑ Aftertreatment controls

4. Technical eld 4: Model-based engine calibration and optimization

5. Technical eld 5: Engine controls modeling

∑ Thermal/uids performance modeling for model-based controls

Table 1.6 Skill sets for each attribute in diesel engine system design

Index Performance Durability Packaging Cost

1 Thermodynamics

2 Combustion

3 Fluid mechanics

4 Heat transfer Heat transfer Heat transfer

5 Tribology Tribology

6 Vibration Vibration Vibration

7 System dynamics System dynamics

8 Controls

9 Acoustics Acoustics Acoustics

10 Probability and

statistics

Probability and

statistics

Probability and

statistics

Probability and

statistics

11 Materials Materials

12 Solid mechanics

13 Cost engineering

14 Finite element

analysis

15 3-D solid modeling 3-D solid modeling

16 Optimization

methods

Optimization

methods

Optimization

methods

Optimization

methods

17 Engine system

design theory

Engine system

design theory

Engine system

design theory

Engine system

design theory

18 Numerical

simulation

analysis

Numerical

simulation

analysis

Numerical

simulation

analysis

19 Engine design Engine design Engine design Engine design

20 Engine testing Engine testing

21 Engine

applications

Engine

applications

Engine

applications

Engine

applications

Diesel-Xin-01.indd 105 5/5/11 11:41:39 AM

