Immune systems inspired approach to anomaly detection, fault localization and diagnosis in complex dynamic systems

Authors:
Dragan Djurdjanovic;Clay Hearn;Yi Liu
Affiliations:
University of Texas, Austin, TX;University of Texas, Austin, TX;University of Texas, Austin, TX
Venue:
Proceedings of the 2010 Conference on Grand Challenges in Modeling & Simulation
Year:
2010

Citing 9
Cited 2

An Intrusion-Detection Model

IEEE Transactions on Software Engineering - Special issue on computer security and privacy
Growing cell structures—a self-organizing network for unsupervised and supervised learning

Neural Networks
Radial basis function based adaptive fuzzy systems and their applications to system identification and prediction

Fuzzy Sets and Systems
Neural Networks: A Comprehensive Foundation

Neural Networks: A Comprehensive Foundation
Divide-and-conquer approach for brain machine interfaces: nonlinear mixture of competitive linear models

Neural Networks - 2003 Special issue: Advances in neural networks research — IJCNN'03
Architecture for an Artificial Immune System

Evolutionary Computation
Issues of Fault Diagnosis for Dynamic Systems

Issues of Fault Diagnosis for Dynamic Systems
Prediction of chaotic time series based on the recurrent predictor neural network

IEEE Transactions on Signal Processing
Prediction of noisy chaotic time series using an optimal radial basis function neural network

IEEE Transactions on Neural Networks

Monitoring of complex systems of interacting dynamic systems

Applied Intelligence
Fault detection and precedent-free localization in numerically discretized thermal-fluid systems

Expert Systems with Applications: An International Journal

Quantified Score

Hi-index	0.00

Visualization

Abstract

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.