In the structural design of pipelines, the currently used methods do not provide a realistic assessment of either the actual safety of pipelines of the reliability of pipe response under a assumed service loading. This may be largely attributable to t...
In the structural design of pipelines, the currently used methods do not provide a realistic assessment of either the actual safety of pipelines of the reliability of pipe response under a assumed service loading. This may be largely attributable to the inadequate determination of design factors that currently enter into calculations are in reality associated with some uncertainty.
In order to study and suggest the design method for underground pipelines based on the reliability analysis several design parameters should be handled, and the design method may be very complicated and unconvenient to use. Among the design parameters, the loads acting on the pipeline due to the backfill are usually being calculated by using the Maston's formula which also contains several design factors. Therefore, it is considered to be reasonable that the loads on the pipeline are taken to be one random variable to simplify the design procedure. The parameters related to the backfill load are considered to be random variables, and random numbers of the parameters are generated. The backfill loads are calculated by using the generated random numbers.
The results are analyzed by the multiple linear regression technique, and regression equations, by which coefficients of variation of the backfill loads can be calculated from the design parameters related to the backfill loads and their coefficients of variation, are proposed and the type of the distribution for the calculated backfill loads is investigated.