Functional Reasoning for Fault Diagnosis Expert Systems

Abstract

Functional Reasoning for Fault Diagnosis Expert SystemsRobert MilneHeadquarters Department of the Army Artificial Intelligence CenterHQDA DAIM -DO, The Pentagon, Wash DC 20310AbstractIndustry today has a severe problem in the automatic testing of analog cards. At the AirForce Institute of Technology, we are developing an Expert System based on the structure andfunction of an analog circuit card to drive automatic test equipment. This system uses theinformation contained in the schematic diagram of the circuit as well as fundamentalknowledge of electronics and past experience in maintaining the card. One of the mostimportant aspects of this system is its ability to reason about possible faults based uponthe function of the sub -sections of the circuit. This task is accomplished using the typeof second principles which an electronic engineer would use. It generates which tests thetest equipment will conduct and, based upon the results of these test, determine the besttest to perform next. In this paper, the basic reasoning mechanism for these systems isdiscussed.IntroductionTesting and fault diagnosis of printed circuit cards is a very important task which isdone many times each day. Typically this is performed using pre- determined, static andrigidly structured tests. As a result, the testing tends to be inefficient, missing manyfaulted components and can often isolate faults to only a large group of components.We introduce the theory of as an approach to automatedtroubleshooting. Using this approach, the understanding of how a circuit works is recordedby assigning responsibilities for parts of the output waveform to subsections of thecircuit.Our current efforts are directed at automated troubleshooting of analog circuit cards.Our overall system, described in [5] and [6], is designed to automatically test and identifyfaults in an analog card through the use of automatic test equipment. Our pastimplementation and papers have conducted testing based on the structural description of thecircuit. In this paper, our approach to functional testing is described.This work differs from others in several significant ways. The work of Cantone [1] hasbeen entirely within the structural area. His algorithm for deciding which test to performbased on the most information gained and possible cost is very helpful to structuralreasoning, but can't help us once we can no longer probe within an sub -circuit. Cantoneonly uses structural information to isolate the fault to a single functional area. In thispaper, we will show how structural information can be used to further propose faults.Dekleer [3] is working to diagnose faults in analog circuit cards from 'firstprinciples'. That is, given the low level electronic description of how a capacitor works,it should be possible to deduce how a filter would work, and consequently, diagnose faults