An approach is presented whereby a number of faults, typical of many hydraulic circuits, may be determined using rule-based concepts. The approach uses on-line data and principally integrates both qualitative and quantitative reasoning in a novel way that enables speciJic faults to be detected. It has the advantage of being able to indicate a class of multiple faults and can also provide some indication of sensor integrity. A practical example is presented using an open-loop meter-inlmeter-out flow control system, and a variety of single and multiple faults have been studied with respect to leakages, check valves andflow control valve behaviour.
Results are presented from a study of leakage losses from a position control systemand the resulting knowledge that may be incorporated into an on-line expert system for fault diagnosis. The use of aminimum number of sensors results in leakage and position error data that have limitations for multiple-faultsituations. It is then shown how the application of an artificial neural network can give additional confidence to a fault assertion, and results are shown for a number of leakage conditions.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.