FIELD: hydro acoustics.
SUBSTANCE: invention relates to underwater acoustics and can be used to build intelligent automated systems for classifying marine targets, detected by the signs of amplitude-phase modulation of low-frequency signals of the pumping of the marine environment by radiation and fields of objects. System for detecting and classifying sea targets using a mathematical model for determining the target type contains a working zone of nonlinear interaction formed in the sea medium and parametric conversion of pumping waves and information waves. Principal difference from prototype is that additionally introduced is an adaptive neuro-fuzzy correction path, comprising a unit of productive rules and functions, the input of which is connected to the output of the neural network recognition and classification channel learning unit, and the output is connected to the input of the neuro-fuzzy network adapter, the function of which is carried out by the adaptive neuro-fuzzy network (ANFIS), and enveloped by the feedback with the differentiator, wherein the output of the neuro-fuzzy network adapter is connected to the input of the differentiator, the output of which is connected to the input of the fuzzy controller, self-adjusting its rule base based on a selection of mathematical models of marine targets, at the output of which the new production rule number signal is generated, as well as a new type of membership function of the target type for the neural network recognition and classification path learning unit, further at the target identification and target recognition unit output by the amplitude-frequency characteristics, neural network recognition and classification channel providing a final classification solution based on detected sea targets, a signal is generated according to the type of target according to the degree of belonging of the analyzed spectrum to the classification object.
EFFECT: automation of the process of recognizing classes of sea targets (surface or underwater object), detected by signs of amplitude-phase modulation of low-frequency pumping signals of the marine environment by radiation and field objects, complex reduction of data size during automatic adjustment of rules base due to generation and reduction of sampling of reference samples of mathematical models of sea targets carried out by means of tract adaptive neuro-fuzzy correction, required for implementation of final classification process in neural network recognition and classification channel, which provides higher probability of correct classification of marine target (surface or underwater object) by 5–7 %.
1 cl, 7 dwg
