Prediction-based approach to determining the metrological reliability of measuring instruments
- PDF / 305,458 Bytes
- 4 Pages / 595.276 x 793.701 pts Page_size
- 65 Downloads / 168 Views
PREDICTION-BASED APPROACH TO DETERMINING THE METROLOGICAL RELIABILITY OF MEASURING INSTRUMENTS UDC 389.14:621
I. A. Sakhnyuk
Abstract. A method is developed for determining the time to metrological failure on the basis of prediction and real models of time distribution of reduced-error values. The proposed prediction-based approach can be used in developing metrological foundations of reliability for measuring instruments. Keywords: metrological reliability, metrological failure, metrological reliability index, time to metrological failure. A topical problem that arises in maintaining measuring instruments (MIs) in world and domestic practice of modern metrology is the determination of their metrological reliability [1]. An analysis of most well-known technical literature sources has shown that the solution of this problem is first of all conditioned by the inapplicability of methods of classical reliability theory to the nature of metrological failures of MIs and the necessity of establishing scientifically grounded verification (calibration) intervals for MIs [2–4]. The objective of this work is the development and application of a new approach to determining the time to metrological failure for MIs on the basis of prediction and real models of distribution of reduced-error values during operating time. The subject of this investigation consists of functional dependencies between metrological characteristics and metrological reliability indices of instruments for redundant measurements that are used to predict time to metrological failure for an MI or the instant of time of its periodic verification. In connection with the variety of modern sensors of physical quantities used in scientific investigations, the problem of metrological reliability acquires a new meaning. First of all, a demand arises for determining the metrological reliability of MIs during their maintenance and developing methodologies, recommendations, and methods that allow one to establish metrologically fault-free states of MIs. The mathematical model of distribution of reduced-error values that is presented in this article with a view to solving the problem of prediction of the metrological reliability of MIs throughout the time to metrological failure is developed on the basis of Weibull distribution function [5–7]. The most significant indicators characterizing the metrological reliability of MIs are the time to metrological failure or metrological resource, metrological safety factor, and instant of time of verification. To achieve the formulated objective, real models of distribution of reduced-error values throughout the entire operating time are constructed on the basis of information on metrological characteristics of MIs (namely, measuring resistance coils) presented by the State Enterprise Ukrmetrteststandart. This information is systematized and a database based on the results of analysis of 174 verification certificates is created. The conventional true valueof the coil resistance R t is found in verifying a measuring resistance
Data Loading...
