Damage Identification with Probabilistic Neural Networks

Efforts to optimize the design of mechanical systems for preestablished use environments and to extend the durations of use cycles establish a need for in-service health monitoring. Numerous studies have proposed measures of structural response for the identification of srrucrural damage, buf few have suggested systematic techniques to guide the decision es to whether or not damage has occurred based on acquired data. Such techniques are necessary because in field applications fhe environments in which sysrems operate and the measurements that characterize system behavior are random. This paper investigates fhe use of artificial neural networks (ANNs) to identify damage in mechanical systems. Two probabilistic neural networks (PNNs) are developed and used to judge whether or not damage has occurred in a speci f ic mechanical syslem, based on experimenral measurements. The first PNN is a classical type that casts Eayesian decision analysis into an ANN framework; ir uses exemplars measured from the undamaged and damaged system fo establ ish whether system response measurements of unknown orjgin come from the former class (undamaged) or the latter class (damaged). The second PNN establishes the character of the undamaged system in terms of a kernel density estimator of measures of system response; when presented with system response measures of unknown origin, it makes a probabilisric judgment whether or nof the dafa come from the undamagedpopu/aCon. The physical system used to carry auf the experiments is an aerospace system component, and the environment used to excite the system is a stationary random vibration. T h e results of damage identification experiments are presented along with conclusions rating the effectiveness of the approaches NOMENCLATURE. ANN : Artificial Neural Network PNN : Probabil ist ic Neural Network PPC : Probabi l ist ic Pattern Classi f ier VETO : Virtual Environment for Test Optimization x y : vector of random variables with dimension n F&J(.) : cumulative distribution function estimator “.