Article 7319

Title of the article

ESTIMATION AND PREDICTION OF THE RESIDUAL RESOURCE BASED ON THE RESULTS OF BINOMIAL TESTS THAT DID NOT GIVE UP 

Authors

Yurkov Nikolay Kondrat'evich, doctor of technical sciences, professor, head of sub-department of radio equipment design and production, Penza State University (440026, 40 Krasnaya street, Penza, Russia), E-mail: yurkov_NK@mail.ru
Mikhailov Viktor Sergeevich, lead engineer, Central Research Institute of Chemistry and Mechanics named after D. I. Mendeleev (115487, 16а Nagatinskaya street, Moscow, Russia), E-mail: Mvs1956@list.ru
Trusov Vasiliy Anatolievich, candidate of technical sciences, associate professor, sub-department of radio equipment design and production, Penza State University (440026, 40 Krasnaya street, Penza, Russia), E-mail: trusov_v@mail.ru  

Index UDK

519.248:62-192 

DOI

10.21685/2307-4205-2019-3-7 

Abstract

Background. For unrecoverable complex products, the GPR does not exceed the minimum operating time of any EREs that make up this complex product, and the probability γ is usually chosen in the range from 0.95 to 0.999. Such a choice of probability values distinguishes the time period of use of the product into intervals, where the initial interval is limited by the value of the gamma-percentage resource (γ≥0.95). This distinction suggests that within this initial interval (15–25 years), the reliability model of unrecoverable complex products is within the influence of an exponential law. This fact allows making predictions of the residual gammapercentage resource of unrecoverable complex products within the established limits (≤25 years). The aim of the work is to find an estimate of the residual gammapercentage resource – within the constraints imposed on the use of the resource, which will be simpler and more efficient than the traditional and inferior to a slightly effective estimate of the residual gamma-percentage resource, if it exists, from the point of view proximity to the true value when using the binomial test plan. And also, based on the results of the extended operation, on the basis of the obtained effective estimate of the residual gamma-percentage resource, obtain a formula for estimating the values of the predicted values of the residual gamma-percentage resource, within the limits of the constraints.
Materials and methods. The study of effective estimates is based on an integral approach, which is based on the construction of a selection rule (criterion) for an effective estimate of the residual gamma-percentage resource given on the sum of the values of the absolute displacements of estimates selected from a certain set.
Conclusions. The obtained estimate of the residual gammapercentage resource is simpler and more effective than the traditional one and is inferior to the slightly effective estimate, if it exists, using the binomial test plan. The resulting estimate of the residual gamma-percentage resource has significant advantages, namely: the estimate is effective on a fairly wide class of biased estimates; assessment allows to obtain the value of the residual gamma-percentage resource from the results of tests that did not give failures. The proposed method of forecasting and the resulting effective assessment of the residual gammapercentage resource have practical application direction in trouble-free operation of products. 

Key words

gamma-percent life, exponential distribution, test plan, point estimation, residual gamma-percentile operating life 

Download PDF

 

Дата создания: 01.11.2019 12:59
Дата обновления: 01.11.2019 13:48