| Summary: | The article presents the problem of the study of developed diagnoses in logic2- and 3-valuable diagnostic system (DIAG 2) devices of the solar power. For this purpose, a functional-diagnostic model of solar power devices has been described. On the basis of the elaborated model of the investigated object, a set of basic elements and a set of diagnostic signals, that are generated by the j-th elements of the functional object, have been determined. Also, there was given a brief description of intelligent diagnostic (DIAG 2) system used for the study. The system (DIAG 2) is a proprietary development of the authors. Diagnostic software in (DIAG 2) system works on the principle of comparison of the set of vectors of diagnostic signals with their standard vectors. By comparing the signals, elementary metrics of vectors disparity of diagnostic signals are determined by the neural network. On the basis of the metrics of differential distance, the system inferences about the diagnosis on the state of elements of a basic object.
Keywords: technical diagnostics, diagnostic reasoning, multivalent logic, artificial intelligence
|
|---|
