Development of a Vehicle Road Load Model for ECU Broadcast Power Verification in On-Road Emissions Testing 2006-01-3392

The 1998 Consent Decrees between the United States Government and the settling heavy-duty diesel engine manufacturers require in-use emissions testing from post 2000 model year engines. The emissions gathered from these engines must be reported on a brake-specific mass basis. To report brake-specific mass emissions, three primary parameters must be measured. These are the concentration of each emission constituent, the exhaust mass flow rate, and the engine power output. The measurement of the concentration level and exhaust mass flow rate can be (and are generally) measured directly with instrumentation installed in the exhaust transfer tube. However, engine power cannot be measured directly for in-use emissions testing due to the direct coupling of the engine output shaft to the vehicle's transmission. Engine power can be inferred from the electronic control unit (ECU) broadcast of engine speed and engine torque. The use of engine power from the ECU presents a problem in that one must rely on the accuracy of the manufacturer's broadcast values. Engine speed is measured directly and is accurate due to the significance of this parameter in the engine control algorithm. ECU broadcast torque is estimated based on models developed during engine development and may be influenced by many parameters that were not accounted for in the model development.

To validate the ECU broadcast power, a vehicle road load model was developed to determine the vehicle's required power. Three primary variables were required for the road load model and include the ECU vehicle speed, the ambient pressure, and time. These three values were measured and recorded and were used in determining the components that describe a vehicle's road load under real-world driving conditions. Although parameters such as driveline efficiency and interactions by the driver could not be accounted for in this study, the results that were determined matched the vehicle's ECU output broadcast power values accurately and were repeatable. Furthermore, the differences between the two different data sets considered provided an estimate for the test vehicle's driveline efficiency and were found to be approximately 82% for the vehicle tested.