PSI - Issue 22

S.V. Belodedenko et al. / Procedia Structural Integrity 22 (2019) 51–58 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

52 2

experts to turn to the method of risk-analysis. The main disadvantage of the classical reliability approach is to take into account failures that have a different scale of consequences. А result of which the reliability function gets a “blurred” character. This leads to a reduction periods between inspections and ov ermaintenance effect. Similar to the reliability methodology, there are two approaches to risk-analysis. In the combined approach (or classical) the properties of the system as a whole are considered. This includes, in particular, the method of F-N curves (Farmer curves). Such models are invariant to the nature (physics) of the damaging process. As described by Fisher et al. (2016), in contrast, in the case of an individual (structural) approach, the risk model takes into account the failure physics. This approach is promising for assessing the technical condition of industrial equipment. The paper was aimed at the development and research of algorithms for assessing the technical condition of mechanical systems according to individual risk indicators of their elements. The task was to illustrate the application of algorithms on real equipment. In risk-analysis, failures are "ranked" depending on the level of loss. Therefore, the predicted operating performance is more accurate. The concept of acceptable risk, on which the theory of industrial safety is based, provides for a staged reassignment of service lifetimes (updating). In case of multifocal damage of equipment, the obtained reliability or risk indices need anyway amalgamated. The amalgamation procedure is actual when using structural reliability methods. This may be in the action of a complex of damaging processes on the structural element, and in the action of a certain damaging process on the system of elements. The importance of amalgamating algorithms in maintenance is due to the fact that the planning of recovery operations takes place, at least, at the machine level. It is inappropriate to appoint inspections for only one element, since at that time the entire system will not be available. The amalgamation of indicators is carried out according to some rules, reviewing which, and developing new ones, is devoted to this report. 2. The amalgamation of the resource safety indices The most commonly used method of amalgamating the individual probabilities of the survival (PS) P i of individual elements of the system into its common PS P Σ is the rule of multiplication:      i n i i I P P 1 . (1) In this case n - number quantity of system elements and degradation processes acting on them. The rule of multiplication of the PS corresponds to the rule of adding individual failures ratio: λ Σ = Σλ i . . Hence for the exponential law of reliability we obtain: exp( ) ЕХР Р t      (2) The widely- known disadvantage of this rule is an excessive drop in the value of Р Σ with an increase in the number of elemen ts of the system (line РI, Fig.1 ). From of one the modifications to the Lindley distribution (Mansour & Hamed, 2015) follows a new rule of amalgamation (Belodedenko et al., 2019): (1 )exp( ) LND P t t        . (3) Marking P EXP = P I , these formulas are transformed into P LND = (1 + λ Σ t) P EXP . Dimensionless risk index of the system is as follows (Belodedenko & Bilichenko, 2015): ik ik Q u     . (4) Taking into account that the deterministic damage a = t / T 0 = λt , the exponential distribution law is transformed