FIELD: computer equipment.
SUBSTANCE: invention relates to a method for finding a faulty unit in a continuous dynamic system based on a sensitivity function. To detect a faulty unit in the system, the monitoring time is pre-determined from an arbitrarily connected dynamic unit, the integral conversion parameter is determined, the number of control points is fixed, the normalized vectors of integral estimates of the output signals of the model are pre-determined in a certain way, the system with the nominal characteristics of the controlled is replaced, similar input signal is fed to input of the system, the integral evaluation of the output signals in the same way as applied to the controlled system is determined, the deviation of the integral evaluation of the output signals of the controlled system is determined, the normalized values of deviations is calculated, the diagnostic signs of the presence of a faulty block is calculated, the structural defect is determined according to the minimum value.
EFFECT: reduced computational costs associated with the implementation of models with structural sensitivity of blocks of transfer functions.
1 cl, 1 dwg
