Planning of the diagnostic experiment in the localization of troubleshooting failures of single-free systems

Formalized conditions for carrying out a diagnostic experiment associated with the identification of faulty fragments (subschemes) of inertial-free systems are obtained.The diagnostic experiment is reduced to computational procedures for localization of faulty subschemes, which are based on testing hypotheses that the characteristics of the allocated subschemes have changed.Hypotheses are formulated in such a way as to ensure the detection of parametric and structural faults.The first, for example, may include changing the resistance of the circuit, and the second -an open or short circuit.When the system is divided into subschemes, the latter are selected from the condition of their parametric identifiably, i.e.The situation when, based on the known parameters of the remaining subschemes, as well as the input and output signals in the whole system, it is possible to determine the parameters of the subscheme under consideration.Planning a diagnostic experiment is as follows.Assuming a linear relationship between the parameters of the allocated subscheme i S and the output signals of the system, a verification matrix F is created in advance, which determines the interrelation of the indicated parameters and signals, considering the serviceability of the subscheme under consideration.In the event of a malfunction, the verification matrix F changes, which allows to determine the nonconformity matrix q , on the basis of which analysis it is possible to obtain an estimate i of the parameters i of the subscheme i S under consideration, as a result of which it is concluded that it is serviceable.Procedures for testing hypotheses on the health of the subschemes of the inertial-free system for cases of parametric and structural faults are of an identical nature.The principal difference between experiment planning and structural faults from planning for parametric faults is that for parametric faults the value i does not depend on the input signals of the system, while for structural faultsthe value i depends on the input signals of the system, in an unknown way.