Energy Efficiency (2019) 12:1053–1064

https://doi.org/10.1007/s12053-018-9717-6

ORIGINAL ARTICLE

A simple approach to estimate ambient temperature bin data

Saban Pusat

Received: 2 February 2017 / Accepted: 15 July 2018 / Published online: 20 July 2018
# Springer Nature B.V. 2018

Abstract In this study, a simple methodology is pro- 2009; Iwaro and Mwasha 2010; Perez-Lombard et al.
posed to estimate ambient temperature bin data. The 2008; Vine 2003), and buildings account for about 45%
proposed model is based on the determination of the best of world total energy consumption (Butler 2008). The
fitting equation describing the characteristics of the cu- increase in population, the increase in comfort level
mulative frequency distribution of yearly bin weather demanded, and the increase in time spent in the buildings
data values. This approach makes it easy for anyone, indicate that the share of buildings in energy consumption
who needs bin data values for any location, to adopt the will increase. Therefore, building energy efficiency stud-
fitted equations in the building energy performance cal- ies become more of an issue.
culations. A case study was applied to six cities in Turkey, Building energy regulations should aim at ensuring
and the applicability of the proposed model has been that people live healthily and safely at the premises while
shown. The obtained R2 values of the fitted equations keeping energy efficiency in the forefront. Estimating the
are higher than 0.99. Therefore, the coefficients of the energy performance of the buildings is important in terms
fitted equations for any location can be used easily to of determining the waste energy and specifying the rea-
predict any bin data value to be used in building energy sons that reduce the energy efficiency. There are lots of
performance calculations. The results of this study are methodologies to determine the energy performance and
important for the experts using bin method. heating/cooling loads of a building. According to the
selected method, it requires little or large amounts of data
Keywords Bin data . Ambient temperature . Heating . and simple or complex calculations (Koulamas et al.
Cooling . Energy 2018). Degree-day method is one of the old, simple, and
fast methodologies used to estimate heating load (Meng
and Mourshed 2017). On the other hand, simulation-
Introduction based dynamic methods like e-Quest, EnergyPlus, ESP-
r, and TRNSYS require detailed data for building and
The rapid increase in world energy consumption is ac- weather conditions (Koulamas et al. 2018). There are also
companied by the depletion of energy resources, energy steady-state methods like BSimplified Building Energy
supply, and environmental problems. An important part of Model^ which ignore dynamic effects to decrease the
energy consumption belongs to building sector (Saidur complexity (Koulamas et al. 2018). Bin and modified
bin methods are also used easily to predict heating and
cooling of buildings (Al-Homoud 2001).
S. Pusat (*) Climate change is a very big concern for the world. It
Department of Mechanical Engineering, Yildiz Technical affects temperatures over the world and predictions for
University, Besiktas, Istanbul, Turkey
the future demands. Brown et al. evaluated the studies on
e-mail: spusat@yildiz.edu.tr
1054 Energy Efficiency (2019) 12:1053–1064

climate change and cooling energy demand for buildings

(Brown et al. 2016). They used cooling degree-day meth-
odology in the analyses. They found the cooling degree-
day balance temperature as 19.4 °C for the USA which is
2 °C higher than the default one, and they defined that the
balance temperatures vary by region.
Auffhammer and Aroonruengsawat investigated the ef-
fects of higher temperatures resulting from climate change
on electricity demand in the residential sector for California
(Auffhammer and Aroonruengsawat 2011). They used bin
methodology in the analyses. In one of the simulations,
they found that total consumption for the households will
rise by up to 55% by the end of the century.
Fazeli et al. evaluated the studies on temperature-based
demand models of energy demand in the residential sector
using basically degree-day method (Fazeli et al. 2016).
They investigated temperature-based models in four sec-
tions as linear symmetric models, linear asymmetric
models, nonlinear models, and semi- or nonparametric
models. They concluded that nonlinear models are more
sensitive to temperature than linear models, and paramet-
ric models are difficult to use and require much data.
Bin method is one of the most commonly used
methods to calculate building heating and cooling energy
loads and to determine HVAC systems such as heating/
cooling coil, fan, chiller, boiler, and cooling tower (Al-
Fig. 1 Summary of the proposed method
Homoud 2001; Li et al. 2014). Bin method uses bin
weather data which consists of frequency of temperature
temperature-dependent behavior and time-dependent pa-
interval occurrence. Part load performance of the build-
ing HVAC systems can be estimated by bin method. rameters (Borge-Diez et al. 2015). They stated that yearly
Many authors have investigated bin weather data and energy use can be predicted by bin method.
Pusat and Ekmekçi prepared bin weather data tables
applied bin method for different locations of the world.
Schicktanz et al. used bin method to calculate solar gains from − 50 to + 50 °C with 3 °C increments in six daily 4-
for heating and cooling applications (Schicktanz et al. h shifts for eight cities representing the distinct climatic
conditions of Turkey (Pusat and Ekmekci 2017). They used
2014). They showed the applicability of bin method in
solar cooling. Wang et al. analyzed energy efficiency of typical meteorological year (TMY) datasets in the calcula-
existing office buildings by using modified bin method tions. Bin weather data tables were generated by Peng et al.
(Wang et al. 2014). They presented a case study for the
model applicability. They concluded that modified bin Table 1 The selected cities for bin data method
method may be used to optimize the energy efficiency
City Station Number Latitude Longitude Altitude
retrofit schemes of existing office buildings and enhance Number °N °E m
the thermal performance. Bulut and Aktacir investigated
the free cooling in İstanbul by using bin weather data and Adana 17,351 37.00 35.20 20
bin method (Bulut and Aktacir 2011). They stated the Adapazarı 17,069 40.46 30.23 31
energy saving potential for transition months (April, May, Ağrı 17,099 39.43 43.03 1631
September, and October), and they found that free Antalya 17,300 36.55 30.48 51
cooling is not useful for hotter months (June, July, and Bartın 17,020 41.37 32.21 30
August). Borge-Diez et al. used bin method to evaluate Bingöl 17,203 38.53 40.30 1177
geothermal source heat pump-building system
Energy Efficiency (2019) 12:1053–1064 1055

Table 2 Bin data for Bartın

Temperature January February March April May June July August September October November December Yearly
bins total

− 50 / − 49 0 0 0 0 0 0 0 0 0 0 0 0 0
− 49 / − 48 0 0 0 0 0 0 0 0 0 0 0 0 0
− 48 / − 47 0 0 0 0 0 0 0 0 0 0 0 0 0
− 47 / − 46 0 0 0 0 0 0 0 0 0 0 0 0 0
− 46 / − 45 0 0 0 0 0 0 0 0 0 0 0 0 0
− 45 / − 44 0 0 0 0 0 0 0 0 0 0 0 0 0
− 44 / − 43 0 0 0 0 0 0 0 0 0 0 0 0 0
− 43 / − 42 0 0 0 0 0 0 0 0 0 0 0 0 0
− 42 / − 41 0 0 0 0 0 0 0 0 0 0 0 0 0
− 41 / − 40 0 0 0 0 0 0 0 0 0 0 0 0 0
− 40 / − 39 0 0 0 0 0 0 0 0 0 0 0 0 0
− 39 / − 38 0 0 0 0 0 0 0 0 0 0 0 0 0
− 38 / − 37 0 0 0 0 0 0 0 0 0 0 0 0 0
− 37 / − 36 0 0 0 0 0 0 0 0 0 0 0 0 0
− 36 / − 35 0 0 0 0 0 0 0 0 0 0 0 0 0
− 35 / − 34 0 0 0 0 0 0 0 0 0 0 0 0 0
− 34 / − 33 0 0 0 0 0 0 0 0 0 0 0 0 0
− 33 / − 32 0 0 0 0 0 0 0 0 0 0 0 0 0
− 32 / − 31 0 0 0 0 0 0 0 0 0 0 0 0 0
− 31 / − 30 0 0 0 0 0 0 0 0 0 0 0 0 0
− 30 / − 29 0 0 0 0 0 0 0 0 0 0 0 0 0
− 29 / − 28 0 0 0 0 0 0 0 0 0 0 0 0 0
− 28 / − 27 0 0 0 0 0 0 0 0 0 0 0 0 0
− 27 / − 26 0 0 0 0 0 0 0 0 0 0 0 0 0
− 26 / − 25 0 0 0 0 0 0 0 0 0 0 0 0 0
− 25 / − 24 0 0 0 0 0 0 0 0 0 0 0 0 0
− 24 / − 23 0 0 0 0 0 0 0 0 0 0 0 0 0
− 23 / − 22 0 0 0 0 0 0 0 0 0 0 0 0 0
− 22 / − 21 0 0 0 0 0 0 0 0 0 0 0 0 0
− 21 / − 20 0 0 0 0 0 0 0 0 0 0 0 0 0
− 20 / − 19 0 0 0 0 0 0 0 0 0 0 0 0 0
− 19 / − 18 0 0 0 0 0 0 0 0 0 0 0 0 0
− 18 / − 17 0 0 0 0 0 0 0 0 0 0 0 0 0
− 17 / − 16 0 0 0 0 0 0 0 0 0 0 0 0 0
− 16 / − 15 0 0 0 0 0 0 0 0 0 0 0 0 0
− 15 / − 14 0 0 0 0 0 0 0 0 0 0 0 0 0
− 14 / − 13 0 0 0 0 0 0 0 0 0 0 0 0 0
− 13 / − 12 0 0 0 0 0 0 0 0 0 0 0 0 0
− 12 / − 11 0 0 0 0 0 0 0 0 0 0 0 0 0
− 11 / − 10 6 0 0 0 0 0 0 0 0 0 0 0 6
− 10 / − 9 3 0 0 0 0 0 0 0 0 0 0 0 3
−9 / −8 7 0 0 0 0 0 0 0 0 0 0 3 10
−8 / −7 5 0 0 0 0 0 0 0 0 0 0 4 9
−7 / −6 4 0 0 0 0 0 0 0 0 0 0 10 14
−6 / −5 11 0 0 0 0 0 0 0 0 0 0 9 20
1056 Energy Efficiency (2019) 12:1053–1064

Table 2 (continued)

Temperature January February March April May June July August September October November December Yearly
bins total

−5 / −4 2 3 0 0 0 0 0 0 0 0 3 15 23
−4 / −3 18 7 2 0 0 0 0 0 0 0 9 16 52
−3 / −2 43 8 9 0 0 0 0 0 0 0 15 37 112
−2 / −1 35 7 17 3 0 0 0 0 0 0 16 33 111
−1 / 0 32 39 14 4 0 0 0 0 0 2 12 77 180
0/1 70 80 20 6 0 0 0 0 0 5 51 91 323
1/2 78 68 56 19 0 0 0 0 0 3 46 61 331
2/3 42 61 61 18 0 0 0 0 0 2 56 39 279
3/4 26 91 88 38 0 0 0 0 0 8 82 61 394
4/5 43 64 133 55 0 0 0 0 0 10 70 36 411
5/6 55 49 117 48 0 0 0 0 0 11 48 47 375
6/7 43 37 79 44 5 3 0 0 2 17 37 49 316
7/8 52 37 41 104 8 2 0 0 9 27 33 20 333
8/9 36 34 12 97 11 6 0 0 7 66 56 25 350
9 / 10 32 14 18 58 16 5 0 0 28 49 28 37 285
10 / 11 30 18 20 37 14 16 0 0 30 56 25 22 268
11 / 12 28 10 16 33 43 33 0 6 52 53 13 12 299
12 / 13 9 18 13 29 72 26 2 10 51 91 22 9 352
13 / 14 9 12 6 20 133 38 8 21 75 76 24 7 429
14 / 15 9 7 8 28 70 47 22 31 69 60 15 4 370
15 / 16 6 3 6 15 61 63 24 33 47 37 16 4 315
16 / 17 3 3 2 13 52 64 41 48 52 37 12 1 328
17 / 18 4 2 1 3 57 72 42 92 46 29 11 5 364
18 / 19 2 0 3 11 37 46 76 106 31 17 10 3 342
19 / 20 1 0 2 2 32 31 82 70 33 20 5 4 282
20 / 21 0 0 0 10 29 36 74 64 32 11 5 1 262
21 / 22 0 0 0 4 32 37 56 37 27 8 0 2 203
22 / 23 0 0 0 6 14 38 50 43 29 8 0 0 188
23 / 24 0 0 0 4 15 41 50 46 17 5 0 0 178
24 / 25 0 0 0 3 11 37 62 34 29 4 0 0 180
25 / 26 0 0 0 2 8 29 70 52 19 9 0 0 189
26 / 27 0 0 0 3 10 20 60 18 16 10 0 0 137
27 / 28 0 0 0 3 5 10 21 17 10 7 0 0 73
28 / 29 0 0 0 0 2 10 3 6 6 2 0 0 29
29 / 30 0 0 0 0 3 5 1 5 3 3 0 0 20
30 / 31 0 0 0 0 1 3 0 5 0 0 0 0 9
31 / 32 0 0 0 0 3 2 0 0 0 0 0 0 5
32 / 33 0 0 0 0 0 0 0 0 0 1 0 0 1
33 / 34 0 0 0 0 0 0 0 0 0 0 0 0 0
34 / 35 0 0 0 0 0 0 0 0 0 0 0 0 0
35 / 36 0 0 0 0 0 0 0 0 0 0 0 0 0
36 / 37 0 0 0 0 0 0 0 0 0 0 0 0 0
37 / 38 0 0 0 0 0 0 0 0 0 0 0 0 0
38 / 39 0 0 0 0 0 0 0 0 0 0 0 0 0
39 / 40 0 0 0 0 0 0 0 0 0 0 0 0 0
Energy Efficiency (2019) 12:1053–1064 1057

Table 2 (continued)

Temperature January February March April May June July August September October November December Yearly
bins total

40 / 41 0 0 0 0 0 0 0 0 0 0 0 0 0
41 / 42 0 0 0 0 0 0 0 0 0 0 0 0 0
42 / 43 0 0 0 0 0 0 0 0 0 0 0 0 0
43 / 44 0 0 0 0 0 0 0 0 0 0 0 0 0
44 / 45 0 0 0 0 0 0 0 0 0 0 0 0 0
45 / 46 0 0 0 0 0 0 0 0 0 0 0 0 0
46 / 47 0 0 0 0 0 0 0 0 0 0 0 0 0
47 / 48 0 0 0 0 0 0 0 0 0 0 0 0 0
48 / 49 0 0 0 0 0 0 0 0 0 0 0 0 0
49 / 50 0 0 0 0 0 0 0 0 0 0 0 0 0

for 26 cities of China (Peng et al. 2009). They prepared characteristics of the cumulative frequency distribution
ambient temperature bin data tables from − 37 to 43 °C of bin weather data values. Therefore, the coefficients of
with 2 °C increments in six daily 4-h shifts. Bin weather the fitted equations for any location can be used easily to
data tables were prepared by Papakostas et al. for 38 cities estimate any bin data value to be used in building energy
of Greece (Papakostas et al. 2008). They prepared ambient performance calculations. The results of this study are
temperature bin data tables from − 18 to 42 °C with 2 °C important for the experts using bin method.
increments in six daily 4-h shifts. Bulut et al. prepared bin
weather datasets for all the Turkish cities from − 36 to 4
5 °C with 3 °C increments in six daily 4-h shifts (Bulut et al. Bin method and bin weather data
2001; Bulut et al. 2003). Bin weather data tables were
generated by ASHRAE for the regions of the USA Bin method is a simple and steady-state way to estimate
(ASHRAE 1995). heating/cooling loads of buildings and energy consump-
However, there is not any study to make use of bin tion of building HVAC systems. Bin data tables com-
weather data tables easier for the experts using these prise the number of hours for the specified temperature
datasets. In this study, a simple and new methodology is and time bins. Hence, bin method is based on the hourly
proposed to estimate ambient temperature bin data measured data. By using bin weather datasets, energy
values. The proposed model is based on the determina- consumption of HVAC systems can be evaluated easily
tion of the best fitting equation describing the by also considering part load performance. For different

Fig. 2 Annual bin data values for 400

Adana
350

300

250
Nbin (h/year)

200

150

100

50

0
-50 -40 -30 -20 -10 0 10 20 30 40 50
Temperature bin (oC)
1058 Energy Efficiency (2019) 12:1053–1064

Fig. 3 Annual bin data values for 500

Adapazarı 450
400
350
300

Nbin (h/year)
250
200
150
100
50
0
-50 -40 -30 -20 -10 0 10 20 30 40 50
Temperature bin (oC)

ambient temperature values, energy load of a building is energy load of the building can be computed with the
calculated with the following equation (Kreider and following equation:
Rabl 1994):
Qtot ¼ ∑m
i¼1 Qbin;i ð2Þ
HB

Qbin;i ¼ N bin;i T b −T o;i ð1Þ where
η
where
Qtot Total energy load (Wh)
m Total number of temperature bins
Qbin,i Energy load (Wh)
Nbin,i Number of hours (h)
HB Total heat loss coefficient for the building (W/oC) Bin weather datasets have been generated for many
ƞ Efficiency of the HVAC system locations around the world to be used in bin method
Tb Balance point temperature (°C) (Degelmen 1985; Jin et al. 2006; Özyurt et al. 2009;
To,i Ambient dry bulb temperature (°C) Papakostas et al. 2007; Papakostas and Sotiropoulos 1997;
Üner and İleri 2000). According to the literature, there is not
The ± sign indicates heating or cooling, i.e., positive a standard for the size of temperature and time bins. In
value of (Tb − To,i) implies heating while negative value general, bin datasets are prepared with 3 °C increments in
indicates cooling. six daily 4-h shifts (Pusat and Ekmekci 2017). Presentation
By using the Eq. 1, energy load of a building can be of yearly bin data values is common (Pusat and Ekmekci
calculated for each ambient temperature bin. Then, total 2017; Peng et al. 2009; Papakostas et al. 2008), but some

Fig. 4 Annual bin data values for 300

Ağrı
250

200
Nbin (h/year)

150

100

50

0
-50 -40 -30 -20 -10 0 10 20 30 40 50
Temperature bin (oC)
Energy Efficiency (2019) 12:1053–1064 1059

Fig. 5 Annual bin data values for 450

Antalya
400

350

300

Nbin (h/year)
250

200

150

100

50

0
-50 -40 -30 -20 -10 0 10 20 30 40 50
Temperature bin (oC)

studies present monthly values (Bulut et al. 2001; Özyurt In bin data studies, the hourly temperature data from
et al. 2009; Papakostas and Sotiropoulos 1997). random years or yearly long-term averages is used.
However, in the present study, the TMY datasets are
used to make the predicted bin data convenient for
The proposed method general use at any location. There are lots of studies
stating that TMY datasets represent the typical climatic
In this section, the proposed bin data estimation meth- conditions of a location (Pusat and Ekmekci 2017; Pusat
od is explained. The proposed method is based on et al. 2015). Therefore, the proposed methodology may
determination of the most appropriate mathematical be applied once to any location for general use.
equation to predict ambient temperature bin data Hourly weather data on a year basis (TMY) are used in
values. For any location, the best fitting model may the proposed method. The proposed method consists of
be found and used in necessary cases. As in the TMY several steps (Fig. 1). Firstly, bin weather data tables are
datasets and bin data tables prepared and published for generated with 1 °C increments from hourly records for
many locations, the best fitting models to predict bin each month. Secondly, start (minimum) and end
data values may be determined, and the coefficients of (maximum) temperature points are determined for yearly
the model may be published. This approach makes it basis. As an extra step for the second stage, some equations
easy for anyone, who needs bin data values of any were fitted to determine the start and end temperature
location, to adopt the fitted equations in the building points by using some basic data of the location. Then,
energy performance calculations. cumulative frequency distributions (CFDs) of outdoor

Fig. 6 Annual bin data values for 450

Bartın
400

350

300
Nbin (h/year)

250

200

150

100

50

0
-50 -40 -30 -20 -10 0 10 20 30 40 50
Temperature bin (oC)
1060 Energy Efficiency (2019) 12:1053–1064

Fig. 7 Annual bin data values for 450

Bingöl
400

350

300

Nbin (h/year)
250

200

150

100

50

0
-50 -40 -30 -20 -10 0 10 20 30 40 50
Temperature bin (oC)

temperature values are prepared. Finally, the best fitting TMY datasets are used to prepare bin data tables. Therefore,
equation describing the characteristics of CFD of yearly the results of this study may be used as a standard conve-
bin data values is determined. niently. TMY method and procedure used in this study
have been explained in the reference (Pusat et al. 2015).
Bin data tables for the selected cities are prepared
Case study for Turkey firstly. In this study, as a rule, bin data tables are pre-
pared for the temperature range of ± 50 °C, with 1 °C
In this section, the proposed bin data method is applied for increments for each month. As an example, bin data
six Turkish cities. The selected cities have different climatic table for Bartın is presented in the Table 2.
characteristics and locations. Some basic data about the Yearly bin data values are calculated and shown in
selected cities are presented in the Table 1. In building the Figs. 2, 3, 4, 5, 6, and 7 for the selected cities. As can
energy performance calculations, both the long-term and be seen from the figures, the shape of the figures chang-
TMY datasets may be used. However, in this study, the es according to the climatic characteristics of the cities.
100

80
Cumulative Frequency (%)

60

40

20

0
-50 -40 -30 -20 -10 0 10 20 30 40 50
Temperature (oC)
ADANA ADAPAZARI AĞRI ANTALYA BARTIN BİNGÖL

Fig. 8 The cumulative frequency distributions of outdoor temperature values

Energy Efficiency (2019) 12:1053–1064 1061

Fig. 9 The cumulative frequency distributions of predicted and measured outdoor temperature values

The highest temperature bins are occurred in Adana Then, the CFDs of outdoor temperature values are
(5 h) and Antalya (1 h) as 41/42 °C. The lowest temper- prepared for the selected cities as a case study by
ature bin is occurred in Ağrı (2 h) as − 33/− 32 °C. using the yearly bin data values (Fig. 8). From the
As a second step in the proposed method, start and end figure, the frequency of ambient temperature values
point temperatures are determined. Temperature varies be- can be determined. As an example, in Ağrı, the fre-
tween − 3 and 42 °C, − 4 and 37 °C, − 33 and 35 °C, − 2 quency of occurring of ambient temperature values
and 42 °C, − 11 and 33 °C, and − 15 and 39 °C for Adana, below the temperature bin of 10/11 °C is 61.55%, and
Adapazarı, Ağrı, Antalya, Bartın, and Bingöl, respectively. the frequency of occurring of ambient temperature
The highest yearly Nbin values are calculated as 385 h (26/ values below the temperature bin of 9/10 °C is
27 °C), 457 h (18/19 °C), 267 h (10/11 °C), 435 h (10/ 58.50%. Therefore, bin data value for the temperature
11 °C), 429 h (13/14 °C), and 426 h (0/1 °C) for Adana, bin of 10/11 °C is calculated as 267.18 h (61.55–
Adapazarı, Ağrı, Antalya, Bartın, and Bingöl, respectively. 58.50 = 3.05%) which is very close to the real value
1062 Energy Efficiency (2019) 12:1053–1064

Table 3 The coefficients and the R2 values for the best fitting R2 value is 0.9992 for Bartın, and the lowest R2 value
equations
is 0.9972 for Bingöl.
City a.x3 + b.x2 + c.x + d R2 In order to use the proposed equations, for example,
one must write nine for x to calculate the frequency of
a b c d occurrence of ambient temperature value below the
temperature bin of 9/10 °C.
Adana − 0.0033 0.1735 0.7978 − 0.1623 0.9987
Due to the change of slope at the beginning and
Adapazarı − 0.0036 0.1340 2.3901 5.7361 0.9984
end of the frequency curves, some erroneous estima-
Ağrı − 0.0007 0.0122 2.3081 39.3810 0.9978
tions are done, which are larger than CFD = 100%
Antalya − 0.0030 0.1517 1.2524 − 0.1222 0.9983
and lower than CFD = 0%. Summary of the erroneous
Bartın − 0.0039 0.1211 2.8434 11.6240 0.9992
estimations are presented in the Table 4. The highest
Bingöl − 0.0010 0.0219 2.5383 26.8340 0.9972
errors are obtained as 176 (over CFD = 100%) and
138 h (below CFD = 0%) for Antalya and Ağrı, re-
spectively. However, the obtained errors are only
(267 h). By using this method, one can compute bin important for higher and lower temperature bins.
data value for any temperature bin. The extra step of the proposed method is applied
Finally, the best fitting equations for the selected for the case study. New equation fitting attempt has
cities are prepared. For each city, linear, second- been done to predict the start and end point temper-
degree polynomial, and third-degree polynomial atures for the prepared bin data tables of the selected
equations are tried, and comparison is made by coef- cities. Many different types of equations (about 500)
ficient of determination values (R2). The average are fitted to estimate the start and end point temper-
calculated R2 values are 0.9716, 0.9787, and 0.9983 atures. The best fitted equation, determined coeffi-
for linear, second-degree polynomial, and third- cients, and R2 values are presented in the Table 5. The
degree polynomial equations, respectively. The pro- proposed equation uses latitude and longitude data of
posed equation for the selected cities is a third-degree the selected cities. As can be seen, R2 values are
polynomial equation. The measured and the predict- higher than 0.98. Therefore, the proposed equation
ed CFDs of outdoor temperature values are presented for the prediction of the start and end point temper-
in the Fig. 9 for each city. As can be seen from the atures can be used conveniently.
figures of each city, it describes the cumulative fre- Therefore, the obtained errors are not important
quency distributions of outdoor temperature values for general use of the proposed method. As a result of
very well. this study, the fitted equations describe the character-
In the Table 3, the coefficients and the R2 of the istics of the cumulative frequency distributions of
fitting equations are given for each city. As can be outdoor temperature accurately and they can be used
seen, the R2 values are larger than 0.99. The highest in bin method easily.

Table 4 The prediction errors for the fitting equations

City Prediction errors

CFD > 100% CFD < 0% Total

Hours Yearly (%) Hours Yearly (%) Hours Yearly (%)

Adana 15 0.17 27 0.31 42 0.48

Adapazarı 112 1.28 25 0.29 137 1.56
Ağrı 95 1.08 138 1.58 233 2.66
Antalya 176 2.01 26 0.30 202 2.31
Bartın 64 0.73 62 0.71 126 1.44
Bingöl 25 0.29 12 0.14 37 0.42
Energy Efficiency (2019) 12:1053–1064 1063

Table 5 The details of the fitting equation for the start and end References
point temperatures

Equation a + b/x1 + c/x2 + d/x22 + e/x23 Al-Homoud, M. S. (2001). Computer-aided building energy
x1 °N Latitude analysis techniques. Building and Environment, 36(4),
x2 °E Longitude 421–433.
ASHRAE. (1995). Bin and Degree hour weather data for simpli-
Start point End point
fied energy calculations. Atlanta: American Society of
Coefficients a − 3774.209 − 1053.630 Heating, Refrigerating and Air-Conditioning Engineers, Inc.
b 1658.1488 2381.0140 Auffhammer, M., & Aroonruengsawat, A. (2011). Simulating the
c 384,018.1 109,867.1 impacts of climate change, prices and population on
d − 13,156,769 − 3,875,652 California’s residential electricity consumption. Climatic
Change, 109(Supplement 1), 191–210.
e 149,799,768 45,250,349
Borge-Diez, D., Colmenar-Santos, A., Pérez-Molina, C., &
R2 0.9977 0.9870 López-Rey, A. (2015). Geothermal source heat pumps
under energy services companies finance scheme to in-
crease energy efficiency and production in stockbreeding
Conclusion facilities. Energy, 88, 821–836.
Brown, M. A., Cox, M., Staver, B., & Baer, P. (2016). Modeling
climate-driven changes in U.S. buildings energy demand.
In the present study, a simple and useful method is pro- Climatic Change, 134(1–2), 29–44.
posed to facilitate of calculation of ambient temperature Bulut, H., & Aktacir, M. A. (2011). Determination of free cooling
bin data values. In the methodology, long-term weather potential: a case study for İstanbul, Turkey. Applied Energy,
data is used to prepare TMY datasets, which are used to 88(3), 680–689.
calculate bin data values, to make bin data convenient for Bulut, H., Büyükalaca, O., & Yılmaz, T. (2001). Bin weather data
for Turkey. Applied Energy, 70(2), 135–155.
general use at that location. The best fitting equations Bulut, H., Büyükalaca, O., & Yılmaz, T. (2003). New outdoor
describing the characteristics of the cumulative frequency heating design data for Turkey. Energy, 28(12), 1133–1150.
distributions of ambient temperature accurately are deter- Butler, D. (2008). Architects of a low-energy future. Nature,
mined in the proposed method. Instead of using bin weath- 452(3), 520–533.
Degelmen, L. O. (1985). Bin weather data for simplified energy
er data tables, the fitted equations can be used in building
calculations and variable degree-day information. ASHRAE
energy load calculations to estimate bin data values. The Transactions, 91(1A), 3–14.
coefficients of the fitted equations may be used easily to Fazeli, R., Ruth, M., & Davidsdottir, B. (2016). Temperature
predict any bin data value to be used in building energy response functions for residential energy demand—a review
performance calculations. The results of this study are of models. Urban Climate, 15, 45–59.
Iwaro, J., & Mwasha, A. (2010). A review of building energy
important for the experts using bin method. regulation and policy for energy conservation in developing
A case study was applied to six cities in Turkey, and countries. Energy Policy, 38(12), 7744–7755.
the applicability of the proposed model has been shown. Jin, Z., Yezheng, W., & Gang, Y. (2006). A stochastic method to
The R2 values of the fitted equations are higher than 0.99. generate bin weather data in Nanjing, China. Energy
Conversion and Management, 47(13–14), 1843–1850.
Therefore, the fitted equations describe accurately the
Koulamas, C., Kalogeras, A. P., Pacheco-Torres, R., Casillas, J., &
characteristics of the cumulative frequency distribution Ferrarini, L. (2018). Suitability analysis of modeling and
of yearly bin weather data values. Additionally, the fitted assessment approaches in energy efficiency in buildings.
equations for prediction of the start and end temperature Energy and Buildings, 158, 1662–1682.
points give high accuracy. As a result of this study, the Kreider, J. F., & Rabl, A. (1994). Heating and cooling of build-
ings: Design for efficiency. New York: McGraw Hill.
proposed method can be used instead of bin data tables Li, Z., Han, Y., & Xu, P. (2014). Methods for benchmarking
for any location, and the fitted equations can be adopted building energy consumption against its past or intended
easily to the building energy performance calculation performance: an overview. Applied Energy, 124, 325–334.
methods. When the TMY data is used in the proposed Meng, Q., & Mourshed, M. (2017). Degree-day based non-
domestic building energy analytics and modelling should
method, the obtained equations are applied continuously
use building and type specific base temperatures. Energy
once they are determined for that location. and Buildings, 155, 260–268.
Özyurt, Ö., Bakirci, K., Erdoğan, S., & Yilmaz, M. (2009). Bin
Compliance with ethical standards weather data for the provinces of the Eastern Anatolia in
Turkey. Renewable Energy, 34(5), 1319–1332.
Conflict of interest The author declares that he has no conflict Papakostas, K. T., & Sotiropoulos, B. A. (1997). Bin weather data
of interest. of Thessaloniki, Greece. Renewable Energy, 11(1), 69–76.
1064 Energy Efficiency (2019) 12:1053–1064

Papakostas, K., Bentoulis, A., Bakas, V., & Kyriakis, N. (2007). Saidur, R. (2009). Energy consumption, energy savings, and emis-
Estimation of ambient temperature bin data from monthly sion analysis in Malaysian office buildings. Energy Policy,
average temperatures and solar clearness index. Validation of 37(10), 4104–4113.
the methodology in two Greek cities. Renewable Energy, Schicktanz, M. D., Döll, J., & Fugmann, H. (2014). Calculation of
32(6), 991–1005. solar gains for solar heating and cooling using the bin-meth-
Papakostas, K., Tsilingiridis, G., & Kyriakis, N. (2008). Bin weather od. Energy Procedia, 48, 1665–1675.
data for 38 Greek cities. Applied Energy, 85(10), 1015–1025. Üner, M., & İleri, A. (2000). Typical weather data of main Turkish
Peng, Z., Jin, Z., Guoqiang, Z., & Yezheng, W. (2009). Generation cities for energy applications. International Journal of
of ambient temperature bin data of 26 cities in China. Energy Energy Research, 24(8), 727–748.
Conversion and Management, 50(3), 543–553. Vine, E. (2003). Opportunities for promoting energy efficiency in
Perez-Lombard, L., Ortiz, J., & Pout, C. (2008). A review on buildings as an air quality compliance approach. Energy,
buildings energy consumption information. Energy 28(4), 319–341.
Building, 40(3), 394–408. Wang, Z., Ding, Y., Geng, G., & Zhu, N. (2014). Analysis of
Pusat, S., & Ekmekci, İ. (2017). Bin weather data for different energy efficiency retrofit schemes for heating, ventilating and
climates of Turkey. International Journal of Global air-conditioning systems in existing office buildings based on
Warming, 12(1), 85–98. the modified bin method. Energy Conversion and
Pusat, S., Ekmekçi, İ., & Akkoyunlu, M. T. (2015). Generation of Management, 77, 233–242.
typical meteorological year for different climates of Turkey.
Renewable Energy, 75, 144–151.