Finding Parametric Defects in Continuous Dynamic Objects Using Isomorphic Models on IIR Filters and Laguerre Filters

The paper considers an algorithm for searching for single parametric defects in continuous dynamic objects using isomorphic models built on recursive IIR filters and Laguerre filters. An algorithm for constructing and tuning isomorphic models is described, and their tuning for the selected diagnostic object is carried out. The sets of vectors of reference defective technical states of the object are calculated. The diagnosis of introduced defects is carried out on the basis of the calculation of their diagnostic features. The accuracy of the algorithm for the two models under consideration is estimated by calculating the statistical coefficients from the confusion matrix and by calculating the values of the identifiability of defects. A comparative analysis of the proposed algorithm with a similar one based on sensitivity models is carried out. Modeling and experiments were carried out using a software package created in Scilab/Xcos.