Particulate matter prediction in both steady state and transient operation of diesel engines

Diesel engines produce a variety of particles generically classified as diesel particulate matter (PM) owing to incomplete combustion. The increasingly stringent emissions regulations require that engine manufacturers must continue to reduce the PM. The ability to predict the PM emissions is one of the key technologies that could be used in a PM reduction strategy. This paper describes a predictive technique that can be used as a virtual sensor for monitoring PM emissions in both steady and transient states for a medium- or heavy-duty diesel engine. The predictive structure is stable over a broad range of engine operation points. The input parameters are chosen on the basis of the PM formation mechanism, physical knowledge of the process, and an insight into the underlying physics. Principal-component analysis (PCA) is used to reduce the dimensionality of the inputs of a non-linear autoregressive model with exogenous inputs (NLARX) from nine inputs to five inputs. PCA not only reduces the input number but also improves the performance of the prediction model. The results show that the NLARX model could predict the particulate matter successfully with an R 2 value above 0.99 with only five inputs.

• Record URL:
  • http://pid.sagepub.com/content/226/2/260
• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/issn/09544070
• Supplemental Notes:
  • Reprinted by permission of Sage Publications, Ltd.
• Authors:
  • Deng, J
  • Bastian, M
  • Stobart, R K
• Publication Date: 2012-2

Language

• English

Media Info

• Media Type: Print
• Features: Appendices; Figures; References; Tables;
• Pagination: pp 260-274
• Serial:
  • Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering
  • Volume: 226
  • Issue Number: 2
  • Publisher: Sage Publications Limited
  • ISSN: 0954-4070
  • EISSN: 2041-2991
  • Serial URL: http://pid.sagepub.com/content/current

Subject/Index Terms

• TRT Terms: Computer models; Data collection; Diesel engine exhaust gases; Diesel engines; Engine operation; Experiments; Fuzzy algorithms; Particulates; Steady state
• Subject Areas: Environment; Vehicles and Equipment; I15: Environment; I90: Vehicles;

Filing Info

• Accession Number: 01363090
• Record Type: Publication
• Files: TRIS
• Created Date: Feb 21 2012 11:12AM