FAULT DIAGNOSIS USING GENETIC ALGORITHMS AND PRINCIPAL CURVES

Abstract

Several applications of nonlinear principal compone nt analysis (NPCA) have appeared recently in proces s monitoring and fault diagnosis. In this paper a new approach is pr oposed for fault detection based on principal curve s and genetic algorithms. The principal curve is a generation of linear princ ipal component (PCA) introduced by Hastie as a para metric curve passes satisfactorily through the middle of data. The exis ting principal curves algorithms employ the first c omponent of the data as an initial estimation of principal curve. However the dependence on initial line leads to a lack of flexi bility and the final curve is only satisfactory for specific problems. In this paper w e extend this work in two ways. First, we propose a new method based on genetic algorithms to find the principal curve. Her e, lines are fitted and connected to form polygonal lines (PL). Second, potential application of principal curves is discussed. An ex ample is used to illustrate fault diagnosis of nonl inear process using the proposed approach.