This report deals with grey box modelling applied to the Well Field Optimisation project. The subject is the real case study of Søndersø, located north-west of Copenhagen, DK. This report contains a comprehensive description on how the dataset of measurements taken at Søndersø have been treated and analysed. The purpose of such analysis is twofold. Firstly is to identify a suitable architecture for the grey-box model. Secondly to design a procedure to select values from the dataset that will be used for the calibration of the parameters of the grey-box model. Section 1 describes the Søndersø well field, and provides an overview of the dataset. Section 2 describes the numeric treatments that have been applied to the dataset; the result is summarized in Section 3. Section 4 illustrates the analysis performed on the treated dataset. In this section, the fundamental mechanisms of the well field system are detected and decomposed (subsections 4.1 - 4.3). Based on the results of such analysis, a simple modelling exercise is performed showing that linear models can be effectively employed to simulate a well field (subsection 4.4). Section 5 describes a sampling approach, designed to calibrate the parameters of the grey-box model with a representative database which is also reasonably reduced in size. Summary and conclusions are in Section 6.
Main Research Area:
Technical University of Denmark, DTU Informatics, Building 321, 2011