Entwurf und Implementation eines Expertensystems für das Detektorkontrollsystem des ATLAS-Pixeldetektors

This thesis describes the design and implementation of an Expert System for the Detector Control System of the ATLAS Pixel Detector at the European center for particle physics, CERN. The ATLAS-Experiment is one of four big experiments installed at the Large Hadron Collider and is itself composed of multiple particle detectors. The innermost of these sub-detectors is the ATLAS Pixel detector, whose more than 30000 operating parameters impose high requirements on the hard- and software of the Detector Control System. In order to support the shift personnel during regular detector operation, an Expert System for the Detector Control System, which provides troubleshooting instructions, was developed within the scope of this work. To allow for a precise mapping of the available human experts knowledge to the Expert System, a rule based and easily adaptable system was designed. The use of the DIM protocol furthermore enables the adoption for other detectors or the data acquisition system. Special emphasis was put on the choice and evaluation of suitable software tools, like the JBoss Rules Expert System shell and the rule syntax to be used. Beyond that, the development of a powerful graphical user interface and a Client-Server architecture were cornerstones in the realisation of the Expert System.