Virtual Sensing: A Neural Network-based Intelligent Performance and Emissions Prediction System for On-Board Diagnostics and Engine Control

Atkinson, Christopher M.; Long, Ting; Hanzevack, E.L.

doi:10.4271/980516

Cited by 66 publications

(32 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Emissions levels are based on the federal transient HD engine certification test 14 and are translated to emissions factors using fuel consumption data. West Virginia University (WVU), through its Transportable Heavy Duty Vehicle Emissions Testing Laboratories (THDVETL), 15 has assembled a bank of data from field 16 relationship for PM emissions from a specific vehicle, the PM must be apportioned continuously over the duration of the test cycle. In this way, PM can then be related to operating parameters such as vehicle speed, acceleration, or load.…”

Section: Objectivementioning

confidence: 99%

Field Measurements of Particulate Matter Emissions, Carbon Monoxide, and Exhaust Opacity from Heavy-Duty Diesel Vehicles

Clark

Jarrett

Atkinson

1999

Journal of the Air & Waste Management Association

View full text Add to dashboard Cite

Diesel particulate matter (PM) is a significant contributor to ambient air PM 10 and PM 2.5 particulate levels. In addition, recent literature argues that submicron diesel PM is a pulmonary health hazard. There is difficulty in attributing PM emissions to specific operating modes of a diesel engine, although it is acknowledged that PM production rises dramatically with load and that high PM emissions occur during rapid load increases on turbocharged engines. Snap-acceleration tests generally identify PM associated with rapid transient operating conditions, but not with high load. To quantify the origin of PM during transient engine operation, continuous opacity measurements have been made using a Wager 650CP full flow exhaust opacity meter. Opacity measurements were taken while the vehicles were operated over transient driving cycles on a chassis dynamometer using the West Virginia University (WVU) Transportable Heavy Duty Vehicle Emissions Testing Laboratories. Data were gathered from Detroit Diesel, Cummins, Caterpillar, and Navistar heavy-duty (HD) diesel engines. Driving cycles used were the Central Business District (CBD) cycle, the WVU 5-Peak Truck cycle, the WVU 5-Mile route, and the New IMPLICATIONS Data comparing integrated opacity and integrated PM mass over a transient diesel chassis test cycle have shown a poor correlation between the two measurements. This implies that opacity measurements may not prove to be a precise tool for identifying high PM emitters. Also, in developing models to describe instantaneous mass emissions of PM from a vehicle, it would be unwise to proportion the measured PM mass over the cycle with respect to opacity readings. There is a stronger correlation between CO and PM measurements for the case of a specific vehicle. Therefore, instantaneous CO readings may be used to proportion PM over a cycle to yield an instantaneous PM value. Even though this method may be undesirably inaccurate, it is the only tool presently available to create continuous PM data for inventory modeling using existing databases.

show abstract

Section: Objectivementioning

confidence: 99%

Field Measurements of Particulate Matter Emissions, Carbon Monoxide, and Exhaust Opacity from Heavy-Duty Diesel Vehicles

Clark

Jarrett

Atkinson

1999

Journal of the Air & Waste Management Association

View full text Add to dashboard Cite

show abstract

“…Starting with two-dimensional graphic representations, other forms like polynomial regressions, neural nets, etc. have been proposed in the literature, see, for instance, [11], [14], [186].…”

Section: Pollutant Formation 271 Introductionmentioning

confidence: 99%

Introduction to Modeling and Control of Internal Combustion Engine Systems

2004

View full text Add to dashboard Cite

SoftcOVel: reprint of the hardcover 1st edition 2004 The use of general descriptive names, registered names trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. Preface Who should read this text?This text is intended for students interested in the design of classical and novel IC engine control systems. Its focus lies on the control-oriented mathematical description of the physical processes involved and on the model-based control system design and optimization. This text has evolved from a lecture series held during the last several years in the mechanical engineering (ME) department at ETH Zurich. The presumed audience is graduate ME students with a thorough understanding of basic thermodynamic and fluid dynamics processes in internal combustion engines (ICE). Other prerequisites are knowledge of general ME topics (calculus, mechanics, etc.) and a first course in control systems. Students with little preparation in basic ICE modeling and design are referred to [52] Why has this text been written?Internal combustion engines represent one of the most important technological success stories in the last 100 years. These systems have become the most frequently used sources of propulsion energy in passenger cars. One of the main reasons that this has occurred is the very high energy density of liquid hydrocarbon fuels. As long as fossil fuel resources are used to fuel cars, there are no foreseeable alternatives that offer the same benefits in terms of cost, safety, pollutant emission and fuel economy (always in a total cycle, or "wellto-wheel" sense, see e.g., [5] and [55]).Internal combustion engines still have a substantial potential for improvements; Diesel (compression ignition) engines can be made much cleaner and Otto (spark ignition) engines still can be made much more fuel efficient. Each goal can be achieved only with the help of control systems. Moreover, with the systems becoming increasingly complex, systematic and efficient system VI Preface design procedures have become technological and commercial necessities. This text addresses these issues by offering an introduction to model-based control system design for ICE. What can be learned from this text?The primary emphasis is put on the ICE (torque production, pollutant formation, etc.) and its auxiliary devices (air-charge control, mixture formation, pollutant abatement systems, etc.). Mathematical models for some of these processes will be developed below. Using these models, selected feedforward and feedback control problems will then be discussed.A model-based approach is chosen because, even though more cumbersome in the beginning, it after proves to be the most cost-effective in the long run. Especially the control system development and calibration processes benefit greatly from mathematical models at early project stages.The appendix contains a brief summary of the most important control...

show abstract

“…Physics-based virtual sensors are discussed by . Overviews on neural network based virtual sensing and sensors are presented by Atkinson et al (1998) and Nareid et al (2005). EGR and residue gas sensing has been researched by Müller et al (2001) andLeroy et al (2009).…”

Section: Development Needs For Engine Sensorsmentioning

confidence: 98%

Diesel engine system dynamics, transient performance, and electronic controls

Xin¹

2013

Diesel Engine System Design

View full text Add to dashboard Cite

Virtual Sensing: A Neural Network-based Intelligent Performance and Emissions Prediction System for On-Board Diagnostics and Engine Control

Cited by 66 publications

References 2 publications

Field Measurements of Particulate Matter Emissions, Carbon Monoxide, and Exhaust Opacity from Heavy-Duty Diesel Vehicles

Field Measurements of Particulate Matter Emissions, Carbon Monoxide, and Exhaust Opacity from Heavy-Duty Diesel Vehicles

Introduction to Modeling and Control of Internal Combustion Engine Systems

Diesel engine system dynamics, transient performance, and electronic controls

Contact Info

Product

Resources

About