Immune Systems Inspired Approach to Anomaly Detection, Fault Localization and Diagnosis in Complex Dynamic Systems
Almost prevalent use of electronics, complicated software, new materials, and technologies makes fault diagnosis and management in contemporary engineering systems increasingly difficult to deal with. Unavoidable design defects, quality variations in the production process, as well as different usage patterns, make it infeasible to foresee all possible faults that may occur on a given system. As a result, traditional precedent-based diagnostic approaches offer a very limited diagnostic coverage based on testing only for the a priori known or anticipated failures, often falsely presuming that the system is operating normally if the full set of diagnostic tests pass. To circumvent these difficulties and provide a more complete coverage for detection and localization of the source of any fault, a new paradigm for design of diagnostic systems is needed. An approach inspired by the functionalities and characteristics of natural immune systems is presented and discussed here. The capability of the newly proposed paradigm to isolate the source of an anomaly without the need to train with signatures characterizing the underlying fault is demonstrated in the simulations of a diesel engine Exhaust Gas Recirculation (EGR) system and a generator portion of a commercially available marine diesel-generator system.