Knowledge-based hydraulic model calibration

Abstract

Model calibration is an essential step in physical process modelling. This paper describes an approach for model calibration support that combines heuristics and optimisation methods. In our approach, knowledge-based techniques have been used to complement standard numerical modelling ones in order to help end-users of simulation codes. We have both identified the knowledge involved in the calibration task and developed a prototype for calibration support dedicated to river hydraulics. We intend to rely on a generic platform to implement artificial intelligence tools dedicated to this task.