Comparison of Predictive Modeling Methodologies for Estimating Fuel Use and Emission Rates for Wheel Loaders

Heavy-duty diesel (HDD) construction equipment consumes significant amounts of fuel and consequently emits substantial quantities of pollutants. The purpose of this paper is to demonstrate three different predictive modeling methodologies for estimating fuel use and emission rates for HDD construction equipment based on real-world, workplace data. Engine performance data for five-wheel loaders, including manifold absolute pressure (MAP), revolutions per minute (RPM), and intake air temperature (IAT) were used to develop prediction models for fuel use and emission rates of nitrogen oxides (NOₓ), hydrocarbons (HC), carbon monoxide (CO), carbon dioxide (CO₂), and particulate matter (PM). For each wheel loader, predictive models were developed using simple linear regression (SLR), multiple linear regression (MLR), and an artificial neural network (ANN). Results indicate that the ANN models accounted for the highest percentage of variability in the data compared with SLR and MLR based on values of the coefficient of determination (R²) for each model. Further, a variable impact analysis was conducted to determine which variables have the most significant effect on fuel use and emission rates for the wheel loaders.

Language

  • English

Media Info

  • Media Type: Web
  • Features: References;
  • Pagination: pp 613-622
  • Monograph Title: Construction Research Congress 2014: Construction in a Global Network

Subject/Index Terms

Filing Info

  • Accession Number: 01525842
  • Record Type: Publication
  • ISBN: 9780784413517
  • Files: TRIS, ASCE
  • Created Date: May 14 2014 3:04PM