| Despite years of environmental concerns and increased regulations our society continues to rely on the fossil fueled internal combustion engine (ICE). The availability and flexibility of ICEs results in new uses on a regular basis. This is especially true in agricultural industries where remote power requirements are common. The first objective of this research was to conduct a customized emissions test and an ISO standard test of a John Deere 4045T diesel engine. The second objective was to create mathematical prediction models based on the two data sets then compare them to determine which test creates better prediction models. The models based on the customized emissions test performed better than the models based on ISO data. The performance based StatNet(TM) models for all three gases were better at predicting emissions than ISO based StatNet(TM) models or models created by the multiple regression formula. Using performance based StatNet(TM) models; CO, HC, and NOx g/hp/hr can be predicted with R2 values of 0.88, 0.96, and 0.89 respectively. |
