1、4784 (RP-1273) The Calculation of Climatic Design Conditions in the 2005 ASHRA E Handbook-Fundamentals Didier J. Thevenard, PhD, PEng Member ASHRAE ABSTRACT ASHRAE Research project 1273-RP recalculated and expanded the tables of climatic design conditions in the ASHRAE Handbook-Fundamentals. These t
2、ables provide values of dry-bulb, wet-bulb, and dew-point temperature, enthalpy, and wind speed at various frequencies of occurrence over annual and monthlyperiods and for some of these, mean coincident values of other variables of interest. Compared to the previous edition of the Handbook, the new
3、tables include additional elements and are calculated for a much greater number of stations over a longerperiod of record. This paper explains the procedure used to compute the design conditions, the data sources used, the techniques employed to screen out erroneous data, and the completeness criter
4、ia required by the calculation. It also provides a summary of the stations included in the 2005 Handbook and a brief description of how the new values compare to thosepublishedin the 2001 edition. Finally, thepaperprovides an overview of the capability of the Weather Data Viewer, a companion CD-ROM
5、that gives full access to the frequency information used to compile the tables of climatic design conditions. INTRODUCTION In support of HVAC design and sizing calculations found throughout its Handbooks (ASHRAE 2001-2004) ASHRAE provides, in a separate chapter of Fundamentals, tables of climatic co
6、nditions for many locations in the United States, Canada, and around the world. These tables include values such as dry-bulb temperature, dew-point temperature, wet- bulb temperature, wind speed and wind direction at various Robert G. Humphries, PhD Associate Member ASHRAE frequencies of occurrence
7、over a long-term period, corre- sponding mean coincident values of some other parameters, and statistics on some extremes. Some tables provide yearly statistics; others provide statistics on a monthly basis. ASHRAE also sells a CD-ROM called the Weather Data Viewer (ASHRAE 2000) that can be used to
8、display actual design values, joint frequency tables, or summary statistics for dry-bulb, dew-point, and wet-bulb temperature, as well as enthalpy and wind speed. In response to the needs expressed by its members and the HVAC community, ASHRAE-sponsored research project 1273-RP aimed at updating the
9、 tables in the ASHRAE Handbook-Funda- mentals to include data for recent years; expanding the number of geographical locations, both in North America and in other countries; including a number of new climatic elements in the tables; expanding monthly percentile tables to include locations outside th
10、e United States; developing a CD-ROM with the entire full-resolution joint frequency information used to compile the design information, and updating the Weather Data Viewer. This paper details how the new tables included in chapter 28 of the 2005 ASHRAE Handbook-Fundamentals (ASHRAE 2005) were deve
11、loped. The paper details the prin- ciples of the calculation method, describes the data sources used, and provides a summary of the information contained in the tables. Didier J. Thevenard is president of Numerical Logics Inc., Vancouver, BC, Canada. Robert G. Humphries is manager of the Air Quality
12、 Modeling and Assessment Group of Levelton Engineering Ltd., Richmond, BC, Canada. 02005 ASHRAE. 457 PRINCIPL SF CALCULATION METHOD Simple Design Conditions Annual simple climatic design conditions included in the 2005 Fundamentals are listed below, with conditions new to the 2005 edition marked wit
13、h an asterisk: 99.6% and 99% heating dry-bulb temperature 0.4%, 1%, and 2% cooling dry-bulb temperature 0.4%, 1%, and 2% evaporation wet-bulb temperature 0.4%, 1%, and 2% dehumidification dew-point tempera- ture and the corresponding humidity ratio “99.6% and *99% humidification dew-point tempera- t
14、ure and the corresponding humidity ratio *0.4%, * 1%, and *2% enthalpy The Handbook also includes the following monthly design conditions: 0.4% and 1% wind speed for the coldest month 0.4%, 1%, and 2% dry-bulb temperature for all months 0.4%, 1%, and 2% wet-bulb temperature for all months The 99.6%
15、and 99% humidification dew-point tempera- tures are used for cold season humidification applications. Enthalpy design conditions are used for calculating cooling loads caused by infiltration and ventilation into buildings. It should also he noted that monthly design conditions werc available only fo
16、r selected US stations in the 2001 Handbook and are provided for all stations in the 2005 edition. The monthly design conditions provide additional information when seasonal variations in solar geometry and intensity, building or facility occupancy, or building use patterns require consideration. In
17、 particular, these values can be used when determining air-conditioning loads during periods of maxi- mum solar radiation. The x% simple design condition is the condition that is exceeded, on average, x% of the time frame under consider- ation. For example, the 1% annual design dry-bulb tempera- tur
18、e is the temperature that is exceeded on average 1 % of the year, or 87.6 hours per year. The 0.4% monthly design wind speed for January corresponds to a wind speed exceeded 0.4% of the time in a typical January month, or roughly three hours during the month. In the previous edition ofthe Handbook (
19、ASHRAE 2001) the calculation of simple design conditions was performed using joint frequency matrices; for example, the 0.4% dry- bulb design condition was calculated by totaling rows in the (dry-bulb, dew-point) joint frequency matrix. This method runs the risk, as will be seen later, that dry-bulb
20、 temperatures are not taken into account for those hours where the dew-point temperature is missing. The method chosen for the 2005 edition was to usefrequency vectors, i.e., calculate the distri- bution functions for individual variables. To calculate simple design conditions for a given variable,
21、long-term hourly data are required. As will be explained later, up to 30 years of data were used. The data are then grouped by bins of equal width. The following bin widths were used for Canadian and international locations: 1 kJkg for enthalpy 0.5“C for dry-bulb, dew-point, and wet-bulb tempera- tu
22、res 1 m/s for wind speed For US stations, the preference was to convert the data back to I-P units (since most were originally recorded in that system of units) and use I-P bin sizes: 0.5 Bhdlb for enthalpy 1 “F for dry-bulb, dew-point, and wet-bulb temperatures 2 mph for wind speed These values are
23、 identical to those used in the 2001 edition. They provide enough resolution while keeping the number of bins relatively low, The frequency vector is defined by counting the number of values in each bin. A cumulative sum of the frequency vector, starting from its lowest value, followed by a division
24、 by the total number of values, provides the cumulative distribu- tion function of the variable. Finally, a lookup of the function enables the determination of the design conditions that are exceeded a given percentage of the time. The procedure will he illustrated with the calculation of the 2% dry
25、-bulb temperature for Atlanta, GA, USA (WMO id: 722 190). Dry-bulb temperature values are sorted by bins 1F (or 0.56“C) wide, leading to the frequency vector shown in Figure 1. Each value of the frequency vector counts all dry-bulb temperatures TDB satisfiing Dry bulb temperature (“C) 10 15 20 25 30
26、 35 40 1,600 T-, -_ 1 O00 . r 0 a 800 r - 600 - 400. 50 59 68 77 86 95 104 Dry bulb temperature (“F) Figure 1 Dry-bulb temperature frequency distribution in August for Atlanta, GA. 458 ASHRAE Transactions: Research where CE represents the center of bin k and ATDB the bin width. The cumulative distri
27、bution function (CDF) is calcu- lated by summing all bins below a certain level and dividing by the total of all bins: i Care should be taken in interpreting the meaning of CDdB. It represents the probability that the dry-bulb temper- ature TDB is less than the upper limit of the bin, that is, ATDB
28、CO$, = PITDB i CE+ - 2 The CDF is shown in Figure 2. The 2% dry-bulb temper- ature is found simply by looking up, on the graph of Figure 2, the dry-bulb temperature corresponding to a CDF of 0.98, or in this case, 923F (333C). Coincident Design Conditions In addition to the simple design conditions
29、described above, the Handbook provides yearly mean coincident condi- tions for a number of variables (conditions new to the 2005 edition are marked with an asterisk): mean wind speed and prevailing wind direction coinci- dent with the 99.6% and 0.4% yearly dry-bulb tempera- tures; mean wet-bulb temp
30、erature coincident with the 0.4%, 1 %, and 2% yearly cooling dry-bulb temperatures; mean dry-bulb temperature coincident with the 0.4%, 1%, and 2% yearly evaporation wet-bulb temperatures; mean dry-bulb temperature coincident with the 0.4%, 1%, and 2% yearly dehumidification dew-point temper- atures
31、; *mean dry-bulb temperature coincident with the 99.6% and 99% yearly humidification dew-point temperatures; mean dry-bulb temperature coincident with the 0.4%, 1%, and 2% yearly enthalpies; as well as the following monthly mean conditions: mean dry-bulb temperature coincident with the 0.4% and 1 %
32、wind speeds, for the coldest month; mean wet-bulb temperature coincident with the 0.4%, 1%, and 2% dry-bulb temperatures, for all months; mean dry-bulb temperature coincident with the 0.4%, 1%, and 2% wet-bulb temperatures, for all months. The calculation of mean coincident conditions requires doubl
33、e binning of the data into what is called a joint frequency matrix. For example, to calculate the mean wet-bulb temper- ature coincident with a given design dry-bulb temperature, one uses a (dry-bulb temperature, wet-bulb temperature) joint frequency matrix. Bin widths ATDE and ATWB are chosen for D
34、ry bulb temperatura (“C) 10 15 20 25 30 35 40 50 59 68 77 86 95 104 Dry bulb temperature (“F) Figure 2 Dry-bulb temperature cumulative distribution function in August for Atlanta, GA. dry-bulb and wet-bulb temperatures, respectively, and element 0, k) of the joint frequency matrix FDB,wB counts the
35、hours during which both of the following conditions are met: ATDE I TDB YDB + - %E 2 2 ATWE *WB 2 5 TwE - 2 (4) (5) where ?iE is the center bin value of dry-bulb bin j, and TkwE is the center bin value of wet-bulb temperature bin k. the mean coincident wet-bulb temperature TkwB,DB can be calcu- late
36、d by simply doing an average of the wet-bulb center-bin temperatures weighted by the values of the corresponding row of the frequency matrix. In mathematical terms: For each dry-bulb temperature center-bin value .i The mean coincident wet-bulb temperature can be calcu- lated for every dry-bulb cente
37、r-bin value, leading to a function such as the one represented in Figure 3. It is then possible to interpolate the function to calculate the mean coincident wet- bulb temperature for any dry-bulb temperature, for example, to calculate the mean wet-bulb temperature coincident with the 2% dry-bulb tem
38、perature calculated in the previous section. This is illustrated again in Figure 3; the 2% dry-bulb temperature being 923F (33.8“C), the mean coincident wet- bulb has a value of 749F (233C). If the weighted averages had been calculated with dry- bulb temperatures over the rows of the matrix, rather
39、than its columns, the results would have been the mean dry-bulb temperatures coincident with wet-bulb temperatures. ASHRAE Transactions: Research 459 ory bulb temperatura PC) 10 15 20 25 30 50 50 59 68 77 ea Dry bulb temperature (F) 35 92 8 F 33 8 C .- u 14 a 12 2 10 95 104 Figure 3 Mean wet-bulb te
40、mperature coincident with dry- bulb temperature in August for Atlanta, GA. Similar procedures can, of course, be used to calculate other mean coincident conditions. The only slightly different case is that of prevailing wind direction coincident with dry- bulb temperature. It is calculated by search
41、ing the maximum cell in a row of the (dry-bulb temperature, wind direction) matrix. Instead of performing a linear interpolation, the value nearest to the considered design dry-bulb value is used. The bin widths used to calculate the mean coincident conditions need not be the same as those used to c
42、alculate the simple conditions. Paradoxically, the use of smaller bins often makes the calculation of coincident design conditions less correct, the reason being that larger bins provide a natural smoothing of the mean coincident function, particularly near the extremes (see Thevenard et al. 2004).
43、The following bin widths were used in the joint frequency matrices for Canadian and international locations: 1 kJkg for enthalpy 10“ for wind direction 1C for dry-bulb, dew-point, and wet-bulb temperatures 1 ms for wind speed For US stations the bin widths were: 2F for dry-bulb, dew-point, and wet-b
44、ulb temperatures 0.5 Btu/lb for enthalpy 2 mph for wind speed 10“ for wind direction The case of the 2F bins for US stations requires special attention. A difficulty arises from the fact that many ofthe dry- bulb and dew-point temperatures were recorded as whole Fahrenheit values. When using 2F bins
45、, this leads to the bins not being properly centered. For example, the bin centered around 40F is expected to hold all values in the interval 39OF, 4 1 “FI; but because whole Fahrenheit values are used, it really contains only 39F and 40“F, so practically it is centered around 39.5“F, not 40F. A car
46、eless use of such 2F bins results in systematic shifts in the calculation of coincident design values, compared to the values that would be obtained with 1 “F bins. The temperature bins were therefore shifted by OSOF, i.e., the bins are centered around 60.5“F, 62.5“F, 64.5“F, etc., instead of 60F, 6
47、2“F, 64“F, etc. Other Design Conditions Other design conditions included in the Handbook include coldest and hottest month, mean and standard devia- tion of extreme annual temperature, monthly mean daily temperature range, and extreme maximum wet-bulb temper- ature, as explained below (conditions ne
48、w to the 2005 edition are marked with an asterisk). *Coldest and Hottest Months. The coldest and hottest months are calculated simply as the month with the lowest or highest average dry-bulb temperature. They may be used, for example, as input to the ASHRAE clear sky model for gener- ation of solar
49、data consistent with the annual heating and cool- ing design conditions. Mean and Standard Deviation of Extreme Annual Dry-Bulb Temperature. The mean TDB,max and the standard deviation of the maximum annual dry-bulb tempera- ture are calculated as N (7) i= I 1 f (DB,mana TDB,max) 2 (8) - i=l ODB,max where N is the number of years for which the maximum annual dry-bulb temperature can be calculated, and is the maximum annual dry-bulb temperature for year i. A simi- lar formula is used for the mean and standard deviation of the minimum annual dry-bulb temperature. *M
