1、278 2009 ASHRAEABSTRACTLife safety can be dramatically improved during a fire by prudent use of pressurization equipment to clear smoke from occupied spaces and escape routes within the structure. Air handling equipment can be utilized to manage and control the route that smoke takes during a fire a
2、nd thereby improve the air quality within the building and at the same time provide greater time for occupants to escape. The issue of life safety is partic-ularly important in ultra high-rise structures that are being planned and are currently being built throughout the world. A smoke control softw
3、are package is used to examine the interacting factors that affect smoke movement in high-rise structures with the overall objective of determining the prac-ticality of using air-handling equipment to pressurize occu-pant spaces in order to keep smoke out and provide a safe area to survive the fire.
4、 Exterior and interior building construction and fire conditions are examined to determine conditions which provide a successful smoke management plan in a high-rise structure. A selective floor pressurization scheme is proposed that reduces the capacity and number of air handling units needed to pr
5、essurize the high-rise structure. INTRODUCTIONLife safety is an extremely important factor in tall build-ings, because many of the traditional approaches to fire safety may not be practical when applied to ultra high buildings. The chimney effect which draws smoke to upper floors is magni-fied in a
6、high-rise building and escape routes are significantly lengthened. These complicating issues, as well as others, require special attention and often un-conventional solutions when selecting a fire safety plan that should be custom-built for a specific building. By intelligent use of equipment to pre
7、s-surize occupant space and escape routes, a fire safety engineer can select a smoke control plan that will optimize all of the factors that are known to influence smoke movement.High-rise buildings create special challenges for fire fighters and for fire protection engineers. Tall buildings neces-s
8、itate extra long escape paths down long stairwells which mean occupants can be exposed to smoky conditions regard-less of where the fire occurs. It may be more practical in high-rise fires to instruct occupants, particularly those on upper floors, to remain in the building and let fire fighting and
9、sprin-kler activity combat the fire. However, this type of approach to a fire safety plan would only be successful if smoke could be kept off all floors in the building with the exception of the fire floor. This type of fire safety plan reduces the necessity of an occupant to descend from upper floo
10、rs and be exposed to smoke along the way, but it places more emphasis on a pres-surization plan to manage smoke movement and keep it away from people who remain in the building. Before a fire safety strategy is devised for a particular building and a given fire scenario, the numerous factors that in
11、fluence the movement of smoke must be understood and quantified. The obvious factors include building design and geometry, weather conditions, location of the fire in the build-ing, fire fighting activities and the intensity of the fire. The shear complexity of the problem suggests that the only pra
12、c-tical way to formulate a rational smoke management plan as a means to improve fire safety is to use a computer program that is based on fundamental principles so that the routes that smoke will take during a fire can be predicted with reasonable certainty. To do so is a formidable task, but once c
13、ompleted, a computer program that can predict the paths that smoke will Pressurization of Floors to Improve Life Safety During a High-Rise FireW.Z. Black, PhD, PEMember ASHRAEW.Z. Black is a regents professor emeritus at the George W. Woodruff School of Mechanical Engineering, Georgia Institute of T
14、echnology, Atlanta, GA.LO-09-024 2009, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transactions 2009, vol. 115, part 2. For personal use only. Additional reproduction, distribution, or transmission in either print or digital f
15、orm is not permitted without ASHRAEs prior written permission.ASHRAE Transactions 279take during a fire can then become a valuable design tool for a fire protection engineer.The smoke control program can evaluate the many options that exist for managing the smoke movement in an effort to address lif
16、e safety issues during a fire. Several of the most obvious schemes would include pressurization of the floor spaces in an effort to provide a safe haven for occupants during the fire. Another option would be to pressurize the fire escape stairwells so that people would have a smoke-free envi-ronment
17、 as they leave the building. Another possibility would be to pressurize one or all of the elevator shafts so that occu-pants could have a safe way to leave the building, particularly those who are handicapped or unable to use the stairwells. A smoke-free elevator would also provide a safe route for
18、fire fighters to ascend to the fire floor with their equipment. This type of plan would also have merit considering the challenge of people attempting to descend from the upper floors in very tall buildings. Any plan to pressurize multiple portions of the building would have to be carefully evaluate
19、d, because to indiscriminately pressurize more than one area within the building could result in the pressurized areas being sources of smoke for areas that are left un-pressurized. The smoke control plan discussed here is based on pressurizing a single area which is the occupied floor spaces within
20、 the building.A smoke control plan based on forcing the smoke to stay within the elevator shafts by pressurizing the floor spaces has several advantages. It does not oppose the natural tendency of smoke to find the path of least resistance to the top of the building. It simply encourages the smoke t
21、o use the large unobstructed elevator shaft as a pathway to the top of the building where it can pass safely to the exterior through the elevator pressure-relief vent. Finally, pressurizing the floors so that smoke remains in the elevator shaft continues the prevailing culture of prohibiting the use
22、 of elevators in the event of a fire. People have long been taught to avoid the use of elevators and the floor pressurization scheme of controlling smoke will simply reinforce and not reverse that long-standing fire safety philosophy.At the outset it is not clear which of the smoke movement options,
23、 or combination of options, will pay the greatest bene-fit. The focus in this paper will be on pressurizing the floor spaces. The other options of pressurizing the stairwells or elevator shafts will be left to future papers.MODEL ASSUMPTIONS, LIMITATIONSAs with all computer models, a number of assum
24、ptions are necessary so that a complex phenomena such as smoke movement can be transformed into a series of mathematical equations. The accuracy of the results is obviously influenced by the assumptions, and therefore the computer output should be viewed with some caution. The program output has bee
25、n thoroughly checked and compared with results from other accepted smoke control models 1, 2. For a common set of input variables including weather conditions, building geom-etry and fan operation, the program provides practically the same results for air flow rates and compartment pressures as the
26、frequently used NIST program called CONTAM. The results are accurate, but of course only within the context of the assumptions used in formulating the model. Since smoke movement is largely governed by the pressure distribution throughout the building, even minute changes in the pressure can have a
27、sizeable effect on the routes that the smoke will take. The effect of fire fighting activities, sprinkler activation, external wind or occupant activities such as opening of stair-well doors can have a significant influence on smoke move-ment, and those effects have not been considered in the formul
28、ation. Therefore the trend in the results is more signif-icant than the actual numerical values of the variables discussed in the following sections. Furthermore, the results shown in the figures are valid only for the specific example being considered and may not extend precisely to other build-ing
29、 geometries or other fire conditions. As much as possible, the discussion of the results will center on the trends caused by changing the parameters and not on the numerical values themselves. The assumptions used in formulating the computer model have been itemized elsewhere 1, 2 and the most funda
30、mental ones are summarized here:The conditions inside the building are steady and do not vary with time. The flow of smoke in the shaft is one-dimensional, vary-ing only with height. There is negligible migration of smoke between floors through the floor slab. The combustion gases and ambient air ar
31、e ideal gases. The intensity of the fire is characterized by the pressure and temperature at the fire floor.The construction within each floor of the building is open, offering no resistance to smoke movement. Large openings in the exterior of the building such as windows and balcony doors remain cl
32、osed. All stairwell doors remain closed on the floors. The temperature change of gases as they pass through openings is negligible.The vertical elevator shaft extends from the fire floor to the top of the building where it vents to the atmosphere and it is not blocked by parked elevator cars. All op
33、enings in the shaft and openings in the exterior surface of the building are uniformly distributed. The model considers the fire floor as the vertical loca-tion of the start of the shaft.The ambient air outside the building has a zero velocity.The cross-section of the shaft is uniform along its enti
34、re height. The air handling units supply air uniformly to each floor.Any attempt to pressurize portions of a building to keep smoke from invading specific areas such as stairwells and floor spaces is greatly dependent on the tightness of those areas. For example, if stairs are pressurized with a ded
35、icated fan system, the capacity of the system used to increase the 280 ASHRAE Transactionspressure above ambient levels is highly dependent on hallway doors remaining closed. An open doorway on any floor can foil a plan to pressurize the shaft and keep smoke from enter-ing the stairwell at that floo
36、r. In a similar fashion, openings on floors due to normal construction openings, unsealed pipe chases, openings for ductwork and all gaps around doors create areas where air at a pressure above the local ambient pressure can escape the floor and complicate any effort to keep smoke from entering that
37、 particular floor. One of the assump-tions in the above list requires that all stairwell doors remain closed throughout the fire. This assumption minimizes open-ings in the structure of each floor. Therefore any calculation of the capacity of the air handling units that are used to pressurize the fl
38、oor space will lead to a volume flow rate that is a mini-mum value necessary to keep smoke from entering that partic-ular floor. If a stairwell door is open on a particular floor, the capacity of the pressurization equipment on that floor must be increased beyond that minimum value to meet the addit
39、ional demand for air on that floor. A single open stairwell door, however, does not influence smoke movement on other floors in a significant way, because floor spaces are fairly well isolated from each other. While a single open door is counter-productive to a floor pressurization fire safety plan
40、on only a single floor, it is far more damaging to an effort to pressurize the stairwells in an attempt to keep them smoke-free. A single open stairwell door can jeopardize the safety of the entire height of the stair shaft, because once smoke enters a stairwell on a single floor through an open doo
41、r, it is able to move rela-tively unimpeded to all floors above. SMOKE MOVEMENT MODELThe smoke movement model used to assess the feasibility of pressurizing floors as a means of improving the quality of air throughout the building has been discussed in some detail in previous papers. The mathematica
42、l foundations have been spelled out in 1, and the program was first used to investigate the potential for pressurizing floor spaces in a high-rise during a fire in 2. The proprietary Fortran program is called COSMO for COntrol of SMOke in a high-rise structure. The program is based upon conservation
43、 of mass, momentum and energy applied to the smoke as it moves upward inside a verti-cal shaft that extends through the entire height of the building. The shaft in this paper is assumed to be a series of elevator shafts. Stairwell doors are assumed to remain completely closed and sealed, and they th
44、erefore do not play a role in the vertical movement of smoke. The smoke control model is an improvement over previ-ous computer models, because it relies upon a detailed heat transfer analysis applied to the energy transfer from the smoke as it moves upward in the shaft. The convective and radiative
45、 heat losses between the hot smoke and the cool walls of the building are used to determine the temperature of the smoke as it moves upward. The detailed heat transfer model is a signif-icant improvement, because the vertical movement of the smoke inside the shaft is driven by the temperature of the
46、 smoke and not its pressure, and an accurate prediction of the smoke temperature is essential to any mathematical model. The program also deviates from other smoke movement programs, because it uses the conservation equations to calcu-late the properties of the smoke. It does not calculate the pres-
47、sure by assuming that the smoke is a static, incompressible fluid like most other smoke movement models. Rather, it determines the pressure by assuming the smoke is a compress-ible ideal gas that is in motion within the building. RESULTSThe number of cases that the program can consider is virtually
48、unlimited. Therefore a systematic approach must be employed to limit the extent of the results and yet provide enough information to draw meaningful conclusions and to identify consistent trends. For this purpose, a set of conditions which are referred to as a “standard” fire and a “standard” buildi
49、ng are identified in Table 1. The selected parameters listed in this table in no way imply that a “standard” building or “standard” fire actually exists; but these terms are used as a simple way to identify a specific building and particular fire conditions and an easy way to identify the large number of input variables to the program. Numerous factors influence the movement of smoke throughout a structure during a fire which makes the task of managing and controlling smoke a very challenging one. Perhaps one of the greatest benefits of a smoke control program like COSMO is its abil
