Validation of a vehicle performance modelling system.

Abstract

Transport costs can account for a significant portion of the total production costs in an agricultural system. In order to ensure maximum efficiency of a transport operation, the transport manager is required to select a suitable vehicle and choose the best route. This is a complex process as vehicles should not be selected based solely on lowest capital cost, but with several variables taken into account. In order to determine an optimal haulage system and to aid the decision making process, a computer model called SimTrans has been developed. SimTrans simulates a haulage vehicle travelling on any given route under normal vehicle operating conditions. The model produces a set of simulated vehicle performance results and a summary of trip details. The vehicle input parameters and route details can be modified to enable the user to simulate different vehicle configurations on alternate routes. The SimTrans model was used in this research project as it provides a comprehensive output of simulated results. However, under certain conditions there is variance between actual and the simulated vehicle performance in SimTrans, particularly with the gear selection routines. The primary objective of this research was to refine, calibrate and validate the current SimTrans model. An initial investigation focused on diesel engine management systems for heavy-duty commercial vehicles and how these systems affect a vehicle's performance. This investigation was aimed at gaining an understanding of how performance data could be gathered directly from the onboard vehicle electronics of a haulage vehicle. A data logging system was developed to measure and record certain operating parameters of a haulage vehicle during operation. The system was installed on a Mercedes Benz 2637 and the vehicle was driven on two alternative routes between the Eston Sugar Mill and a loading zone near Mid-Illovo. The Mercedes Benz 2637 used for the vehicle tests was simulated in SimTrans using identical operating conditions of route and speed limits. By comparing the observed and simulated data sets, errors particularly with gear selection routines were identified. The SimTrans model was modified such that these errors were corrected and a new gear skipping routine was added. The modified SimTrans model was validated through a comparison of a new set of simulated results with the observed data. The modified model showed significant improvement in simulation accuracy over the previous version of SimTrans.