Page 332 - Pipeline Risk Management Manual Ideas, Techniques, and Resources
P. 332

Hazard zone calculations 14/309
                People located indoors at the time of failure would likely be   flammable limit is often chosen for hydrocarbon gases. The use
                afforded indefinite protection and       of the lower flammability limit-the  minimum concentration
                People  located  outdoors  at  the time  of  failure  would  be   of gas  that  will  support  combustion-is   the  most  common
                exposed to a finite but low chance of fatality [83].   cloud boundary. It conservatively represents the maximum dis-
                                                         tance from the leak site where ignition could occur. Sometimes
                If 5000 Btu/hr-ft2 is used then the previous equation  (for   112 of the LFL is used to allow for uneven mixing  and the
              methane) simplifies to:                    effects of random cloud movements. This lower concentration
                                                         creates  a  larger cloud. In the case of a toxic  gas, the cloud
                             r=0.685 x JpXdZ             boundary  must  be  defined  in terms of toxic  concentrations.
                where                                    Note that unignited sour gas (hydrogen sulfide, H2S) releases
              r = radius from pipe release point for given radiant heat inten-   have been estimated to cause potential hazard zones 4 to  17
                 sity (feet)                             times greater than from an ignited release [95].
              p = maximum pipeline pressure (psi)          The extent and cohesiveness of a vapor cloud are critical
              d = pipeline diameter (inches).            parameters in determining possible threats from that cloud, as
                                                         is discussed  in Chapter 7. Meteorological  conditions have a
                Note that thermal radiation intensity levels only imply dam-   large influence on these parameters. In most dispersion mod-
              age states. Actual damages are dependent on the quantity and   eling, an atmospheric stability class is often assumed as part
              types of receptors that are potentially exposed to these levels.   of the model requirements. This can be based on analyses of
              A preliminary assessment  of structures has been performed,   weather patterns in the area of interest or simply defaulted to
              identifying the  types  of  buildings  and  distances from  the   conservative (worst case) conditions. Often, an atmospheric
              pipeline.                                  class F-moderately   stable-is   chosen because it results  in
                                                         larger  cloud  sizes compared to conditions  involving  more
              Pool fire damages                          wind  or  sunlight  effects. The  information  shown  in  Table
                                                         14.31 is often used in determining the atmospheric stability
              As an example of some hazard zone distances for a delayed-   class.
              ignition, zero wind gasoline pool fire event, the information in
              Table  14.29 was extracted from published work  done  in the   Vapor cloud explosion
              United  Kingdom  [58]. Details  of  the  calculation  procedure
              were not presented in this reference. Note that the pool diame-   The mechanisms thought to underlie the detonation of a vapor
              ter is the most critical factor in most calculation procedures (see   cloud are generally discussed in Chapter 7. This event poten-
              Chapter 7). Therefore, factors such as release rate, topography,   tially occurs as a vapor cloud combusts in such a rapid manner
              and soil permeability  are needed to estimate pool size. Table   that a blast wave is generated. A confined vapor cloud explo-
              14.30 is another example of gasoline pool hazards. This table   sion is more common than unconfined, but note that even in an
              also shows hazards from oil pools for comparison. While haz-   atmospheric release, trees, buildings, terrain,  etc., can create
              ard distances are similar for oil and gasoline, note the signifi-   partial confinement conditions. Any explosive event can have
              cant differences in ignition probabilities between the products.   associated  missiles  and  high-velocity  debris  whose  damage
              Other examples of hazard zone distances can be found in the   potentials have been dramatically demonstrated but are very
              case studies later in this chapter and in examples of spill scoring   difficult to accurately model.
              shown in Chapter 7.                          The explosive potential  of a vapor  cloud  is related  to  its
                                                         mass, heat of combustion, and the amount of total energy that
              Vapor dispersion                           would  contribute  to  the  explosive  event-the   yield  factor.
                                                         Yield factors are critical for the calculation, but are the least
              As discussed  in Chapter 7, the release of a gaseous pipeline   precisely  known variable  in the  calculation. They  generally
              product  creates  a  vapor  cloud,  the  extent  of which  can  be   range from 2 to 20% and some representative values are shown
              defined by some concentration of the vapor mixed with air. A   inTable 14.32.

              Table 14.29  Sample hazard zone distances
              ~~~
              Hole diameter   Soil type   Release mte (kb/sj   Pool diameter (m)   Flame length (m)   Radial distance to 10 k W/m2 (m)
               16 in. (406 nun)   Average   205        100           78              126
                              Clay
               12.75 in.1324 mni)   Average   164      100           78              I26
               8.625 in. (219 mm)   Average   100       85           70              110
                              Clay         I00         100           78              I26
              6.625 in. (168 mm)   Average   30         46           46               65
               1 O-mm leak    Average       5.3         19           25               30
                              Clay          5.3         13           63               96
              Source: Morgan, E., "The Importance of  Realistic Representation of Design Features in the Risk Assessment of  High-pressure Gas Pipelines," pre-
               sented at Pipeline Reliability Conference, Houston, TX, September 1995.
              Note: Diameters shown are maximum spreading pool diameters reached before ignition. The diameters have been limited to 100 m maximum.
   327   328   329   330   331   332   333   334   335   336   337