Page 172 - Reliability and Maintainability of In service Pipelines
P. 172

Case Studies on the Application of Structural Reliability 157


                      1
                     0.9
                     0.8
                   Probability of failure  0.7       Monte Carlo

                     0.6
                     0.5
                                                     First passage (ρ = 0.5)
                     0.4
                     0.3
                     0.2                             GDD model  d
                     0.1
                      0
                        0    25    50   75   100   125   150  175   200
                                           Time (year)

           Figure 5.25  Verification of the results from the two methods by Monte Carlo simula-
           tion method.

           Sensitivity Analysis
           For a comprehensive reliability analysis it may be appropriate to assess the effect
           of different random variables on the service life of the sewer. It is of interest to
           identify those variables that affect the wall thickness reduction most, so that further
           studies can focus on those variables. For this purpose, relative contributions of vari-
           ables to the variance of the limit state function were calculated using Eq. (3.8).
                                           2
             Fig. 5.26 illustrates sensitivities (α ) or relative contribution of variables to
           the variance of the limit state function calculated based on the definition in
           Eq. (3.8) (Section 3.5). It can be concluded that while the contribution of some
           variables are relatively small, some other variables have considerable contribution
           to the limit state function. High values of contribution of variables mean that the
           sensitivity to those variables is more dependent on the actual value of their coeffi-
           cient of variation. In such cases, more concern should be taken in order to deter-
           mine relevant parameter values.
             The GDD model was run to elaborate on the effect of variables with high

           values of relative contribution (i.e., [DS], A, and b=P) on the probability of fail-
           ure. As can be concluded from Fig. 5.27, a change in sulfide concentration has a
           considerable effect on service life of the concrete sewer.
             Fig. 5.28 also shows the effect of sewage flow quantity on the service life of
           the sewer. The ratio of the width of the stream surface to the perimeter of the

           exposed wall (b=P) has a significant effect on the corrosion rate of concrete and
           consequently the service life of the sewer decreases dramatically by increasing

           the ratio of b=P.
             To investigate more about the effect of concrete properties on the service life
           of concrete sewers, more studies on the effect of change in alkalinity (acid
   167   168   169   170   171   172   173   174   175   176   177