Features of assessing the condition of complex objects

Abstract

In practice, complex objects are characterized by several parameters that can be correlated. Such facilities include most of the facilities / subjects of diagnosis in medicine, ecology, biology, meteorology, climatology, in assessing the quality of raw materials, all types of fuels and others. Assessment of the state of complex objects is carried out by ensuring the maximum completeness of the information about the parameters of the studied object by measuring the parameters that correlate with the studied parameter. The degree of commonality of the processes of measurement and diagnosis in terms of information is significant, which leads to the need for their joint study and to obtain relationships that are fair in both measurements and control, which increases the reliability of diagnosis. Since the controlled parameters, in the general case, can change over time, it is necessary, based on the assumption of the influence of only random variables, to introduce the norms of possible deviations. Going beyond these norms is the basis for decisionmaking. However, independent monitoring of individual indicators may, if correlated, lead to erroneous decisions. To prevent this, it is proposed to use statistical methods with multiparameter criteria to assess and diagnose the stability of complex objects. The paper proposes the use of T ^ 2 - Hotelling statistics to diagnose the state of complex objects. The physiological state of the human body was chosen as the object of study.