Determination of In-Use Brake-Specific Emissions from Off-Road Equipment Powered by Mechanically Controlled Diesel Engines

Exhaust emissions from off-highway diesel engines are a significant contributor of both oxides of nitrogen (NO_x) and particulate matter (PM) to air inventories. Yet, emissions research activities aimed solely at the off-highway arena have been minimal - largely overshadowed by the extensive efforts directed toward the on-highway sector. However, with current trends indicating that the performance of these off-highway vehicles will become increasingly more scrutinized by federal regulatory agencies, augmentation of current research efforts will be necessary.

The global objective for this study was to collect vehicle activity information for diesel-powered off-highway vehicles while they were operated in the field. Engine speed and raw exhaust CO₂ concentrations were recorded and then used to create engine dynamometer test cycles. The engine was exercised according to these cycles in the laboratory so that the mass emissions rates of exhaust gas pollutants could be measured. To determine if the in-field raw exhaust CO₂ and engine speed measurements could be used to accurately predict engine power output, these quantities were recorded operating the engine over a series of load maps at predetermined engine speed ranges on a dynamometer test bed. The resultant data was used to develop CO₂ vs. engine output power performance maps that could be used to infer engine power from raw exhaust CO₂ concentration and engine speed measurements. The engine was then exercised according to a step input cycle while raw CO₂ emissions data were collected so that an estimate of the engine loads encountered during the running of the step input cycle could be formulated. Results showed that this surrogate method of horsepower estimation was capable of providing results that were within 5% of actual measured values for steady-state engine operating points. The accuracy of predicting engine power output from measured raw CO₂ concentrations and engine speed was further improved by comparing the in-laboratory transient cycle emissions data with the in-field raw CO₂ exhaust concentration data and adjusting cycle setpoints in an iterative fashion until satisfactory correlation existed between the in-field raw exhaust CO₂/engine speed traces and the raw exhaust CO₂/engine speed traces measured in the laboratory. The converged test cycle setpoints were then used to exercise the engine so that exhaust mass emissions measurements could be made - thus providing an estimate of the actual in-field, in-use mass emissions produced by the off-road test vehicle. The effect of cycle recreation accuracy upon emissions results was also investigated by comparing the engine emissions results from different stages in the iteration process.