A Framework for Semi-Automated Generation of a Virtual Combine Harvester

This paper describes a generic data-driven model of the threshing, separation and cleaning process in a combine harvester. The aim is a model that describes the actual material flow and sensor values for relevant actuator configurations and measured environmental disturbances in order to facilitate Hardware In the Loop (HIL) simulation and sensor based material flow estimation.

A modular data-driven model structure is chosen as it maintains the actual steady-state values and facilitates verification and debugging using laboratory and field data. The overall model structure, model generation procedure, and estimation of parameters from field data are described, as well as simulation results are presented.