We propose a novel framework for Model-Based Diagnosis (MBD) that uses active testing to decrease the diagnostic uncertainty. This framework is called LYDIA-NG and combines several diagnostic, simulation, and active-testing algorithms. We have illustrated the workings of LYDIA-NG by building a LYDIA-NG-based decision support system for the Gravity field and steady-state Ocean Circulation Explorer (GOCE) satellite. This paper discusses a model of the GOCE Electrical Power System (EPS), the algorithms for diagnosis and disambiguation, and the experiments performed with a number of diagnostic scenarios. Our experiments produced no false positive scenarios, no false negative scenarios, the average number of classification errors per scenario is 1.25, and the fault detection time is equal to the computation time. We have further computed an average fault uncertainty of 2.06 × 10−3 which can be automatically reduced to 9.5×10−4 by sending a single, automatically computed, telecommand, thus dramatically reducing the fault isolation time.
Model-Based Diagnostic decision-support system for satellites
2013-03-01
497874 byte
Conference paper
Electronic Resource
English
Model Diagnostic Analysis of Global Coverage Satellites
AIAA | 2014
|Model Diagnostic Analysis of Global Coverage Satellites (AIAA 2014-4462)
British Library Conference Proceedings | 2014
|AI based diagnostic and decision support systems for the trucking industry
Automotive engineering | 1990
|Satellites support train order working
IuD Bahn | 1995
|