ADAPTIVE BAYESIAN MODELS FOR EVALUATING THE INDICATORS OF THE RELIABILITY OF FREIGHT CARS ON THE STAGES OF THE LIFE CYCLE

Abstract

The article discusses the rationale for the application of the Bayesian model (BM) for calculating the reliability indicators of a freight car at the stages of its life cycle. The life cycle includes such stages as the creation, production, operation, repair and utilization of the car. Each of the stages, in turn, consists of several operations and procedures. So, for example: the design phase begins with pre-design solutions and ends with the creation of the car; the production phase - from production to decommissioning, and the operation and maintenance phase - from the start of operation to the disposal of the car. BM allows to evaluate a posteriori (after the implementation of actions) probable hypothesis, if the facts on the implementation of certain actions have become known. The work investigated the fulfillment of requirements for incompatibility of action groups (controlled processes), the data of which are used in models of adaptation of system parameters that control the reliability indicators of cars. A freight wagon is considered, in the nodes (wheel sets) of which mechanical wear and fatigue damage occur. For the study, the mileage of 160 thousand km was taken into account, since within these limits the indicated damages are independent, and the approximation error is an acceptable 16 %, which ensures the application of the Bayesian approach. On the basis of experimental data, indicators of reliability of parts and components of freight cars were determined. The calculated reliability indicators and the developed individual scheme of the freight car model allowed us to construct the failure-proof matrix of individual elements and nodes of gondola cars 12-7023 for different operating times (runs). Analyzing the elements of the matrix, the vulnerable parts and assemblies of the gondola cars were identified, changes in the design of the cars were recommended, the overhaul run or repair time was increased. The usage of adapted BM models according to the presented methodology allows assessing the reliability indicators at the stages of the life cycle, optimizing the turnaround time and number of repairs during the life cycle of the car model, while ensuring that structural features are taken into account, as well as the probabilities of failure-free operation of the wagon-bearing elements and components.

Authors and Affiliations

Sergiy Miamlyn, Vladislav Skalozub, L. Muradian

Keywords

Related Articles

FORMATION OF THE METHOD FOR SELECTION OF VEHICLES LOAD CAPACITY TO WORK WITH SCHEMES OF CONVEYING UNITIZED CARGO

The article is devoted to the research of the method for selection of vehicles load capacity to work with schemes of conveying unitized cargo. The first stage was to carry out simulation of conveying unitized cargo at di...

DESIGN OF POLYGONAL COVERAGE COATINGS OF WOOD

Currently in Ukraine the production of wooden I-beams with a wall of OSB is a steady increase. Analysis of the structural forms of composite beams, the features of their use in the frame of buildings, along with the exis...

MODELING OF RANDOM LOADS ON THE CONSTRUCTION OF FRAMEWORK BUILDINGS

Determination of real indicators of the reliability of structures in framework buildings is a rather complicated task. The solution of this problem is possible only on the basis of an accurate mathematical description of...

INFLUENCE OF SLAGS OF BOILERS WITH CIRCULATING FLUIDIZED BED ON FROST RESISTANCE OF CONCRETE

The results of the study of the influence of slags of boilers with a circulating fluidized bed on the frost resistance of heavy concrete are presented. In the studies was used fallowing materials: Portland cement PC I-50...

TECHNOLOGICAL SOLUTIONS FOR OPTIMIZATION OF MATERIAL AND LABOR COSTS AT THE ANKEROUS INSTALLATION WORKS WITH THE USE OF ACRYLIC COMPOSITIONS

The results of the study of the influence of organizational and technological factors on the duration of installation in concrete of chemical anchors using acrylic compositions and the conditions for the execution of wor...

Download PDF file
  • EP ID EP536329
  • DOI 10.18664/1994-7852.182.2018.160515
  • Views 69
  • Downloads 0

How To Cite

Sergiy Miamlyn, Vladislav Skalozub, L. Muradian (2018). ADAPTIVE BAYESIAN MODELS FOR EVALUATING THE INDICATORS OF THE RELIABILITY OF FREIGHT CARS ON THE STAGES OF THE LIFE CYCLE. Збірник наукових праць Українського державного університету залізничного транспорту, 182(16), 80-91. https://europub.co.uk./articles/-A-536329