Construction of separating function for objects classification and its use for classification of spherical segments from material with shape memory

An engineering method for constructing a separating surface for discriminating classes of multiparameter objects is described. The method is based on the proposed approximation of the membership function, which makes it possible to classify with a complex form of the separating surface. An algorithm is given for calculating the estimates of the coefficients of the nonlinear model with respect to the coefficients of the separating surface model using an iterative procedure based on the method of least squares. The results of experiments and calculations on the classification of spherical segments from materials with shape memory, proving the effectiveness of the proposed method are presented. (paper)