Building and Environment 39 (2004) 959–967www.elsevier.com/locate/buildenv

Residential end-use energy simulation at city scale

Yoshiyuki Shimoda∗, Takuro Fujii, Takao Morikawa, Minoru MizunoDepartment of Environmental Engineering, Graduate School of Engineering, Osaka University, 2-1 Yamada-oka, Suita Osaka 565-0871, Japan

Abstract

This paper describes a simulation model for predicting end-use energy consumption in residential sectors of a city or region. In thismodel, the annual energy consumption of a dwelling is simulated from the occupants’ schedule of living activities, weather data andenergy e2ciencies of appliances and dwellings. By summing up the simulation results for various household categories, total energyconsumption for the residential sector in a region can be estimated.

In this paper, energy consumption for Osaka City is simulated. The result is compared with statistical data. The e5ects of energye2ciency standards and urban heat island phenomena are examined.? 2004 Elsevier Ltd. All rights reserved.

Keywords: Residential end-use energy; City-scale simulation; Energy e2ciency standards; Urban heat island

1. Introduction

Energy consumption in Japanese residential sector is in-creasing continuously. In the last 25 years, energy consump-tion of residential sector has been doubled while the pop-ulation has increased by only 10%. This is because of thechange of life-style, increase of home electric appliances,tendency to increase the number of families with small mem-bers, enlargement of dwellings, improvement in indoor ther-mal environment and so on. To accomplish national en-ergy and carbon dioxide reduction targets, various kindsof measures, such as improvement in thermal insulation ofdwellings, improvement in energy e2ciency of home elec-tric appliances, introduction of renewable energy system andintroduction of residential combined heat and power genera-tion system have been proposed. However, city-scale e5ectsof these measures are not clear since energy consumptionof each household di5ers considerably depending on house-hold type, dwelling type and other factors. Especially, sinceintermittent air-conditioning is common in Japan, the dif-ferent schedules in living activities of occupants a5ect theamount of energy consumption and e5ects of the energysaving measures greatly. Therefore, simulated e5ects of theenergy saving measures for ‘the standard household’ which

∗ Corresponding author. Fax: +81-6-6879-7665.E-mail address: shimoda@env.eng.osaka-u.ac.jp (Y. Shimoda).

means a couple and two children, is not applicable for cityor national scale evaluation.In addition, since an increase in the number of elderly

people and changes in the composition of household typeare expected in the near future, a consideration for the com-position of households is essential to simulating city-scaleenergy consumption in the future’s residential sector. En-ergy e2ciency of the stock of dwellings and appliances isalso important to estimate present and future energy con-sumption precisely.Jones et al. [1] estimated city-scale energy use consider-

ing the di5erence of house characteristics, but energy con-sumption by each appliances are not calculated. Michalik etal. [2] developed structure model of energy demand in theresidential sector of a region by summing up each appli-ances’ operation schedule. However, this method does notinclude heat load calculation.In this paper, we developed a model which simulates

city-level energy consumption in the residential sector con-sidering the composition of household type and dwellingtype. By applying this model for Osaka City (Population:2600 thousand, 1058 thousand households), the presentamount of energy consumption in the residential sectoris estimated and compared with statistics. In this calcula-tion, all of the households in Osaka City are classiGed byhousehold type and dwelling type. After that, hourly energyconsumption is simulated for each household category.This simulation model consists of appliances’ operation

0360-1323/$ - see front matter ? 2004 Elsevier Ltd. All rights reserved.doi:10.1016/j.buildenv.2004.01.020

960 Y. Shimoda et al. / Building and Environment 39 (2004) 959–967

schedule model, hot water supply model, heat load calcu-lation model and heating/cooling equipment performancemodel. By adding the scenario of how the dwelling and ap-pliance stock will be changed and how the population andthe composition of household will be changed, the model isable to quantify the impact of each scenario on total energyconsumption in Osaka City’s residential sector.In the last part of this paper, the impact of new energy

e2ciency standard for thermal insulation of dwelling androom air conditioner on total energy consumption in OsakaCity is evaluated. The inJuence of urban heat island phe-nomena is also estimated.

2. Simulation method

2.1. Simulation procedures

Fig. 1 shows the Jowchart for the simulation. In thissimulation, energy use of dwellings is calculated iterativelyfor 23 household types and 20 dwelling types (10 types fordetached house and 10 types for apartment house). Thenannual energy consumption for every one hour for 460types of dwellings is calculated. Occupants schedule ofliving activities, amount and temperature of hot water sup-ply, set temperature of cooling and heating, weather data,appliances’ energy performance properties and dwellings’thermal properties are provided as input data for thissimulation.By multiplying simulated energy consumption and the

number of households for each category and then summing

Weather Data

Total Energy Consumption in the Area

The number of households classified bytypes of household members and floor

area of dwellingClassified into 23 household types, 2 dwellingtypes, 10 floor area categories from National

Population Census 2000

Dissemination ratio and energyefficiency of each types of appliancesAverage thickness of insulation

Energy use pattern modelEnergy use for home electric

appliances, lighting and cooking are simulated from occupants’ living activities at each time step

Hot water supply modelEnergy use for hot water supply is calculated from the amount of hot water use and temperature of city water and hot water.

Cooling and Heating modelDynamic cooling and heating loadcalculation and energy consumption of room air conditioners and other heating appliances

Energy consumption for each household typeEvery 15 min., type of household, type and area of dwelling.

The amount and temperature ofhot water use Set temperature of cooling and heating

Input data setEnergy simulation for one dwelling

Type of household and dwelling

23 household types and 20dwelling types (10 types of detached house + 10

types of apartment house)

Occupants’ schedule of livingactivities

(National time use survey, NHK)

Internal heat gain

Stock Model(in developing)

365daysEvery 1 hour

365daysEvery 15 min.

3 season (summer, winter, intermediate)

Weekday/holidayEvery 15 min.

Fig. 1. Flowchart of the simulation.

them up, total energy consumption for the residential sectorin the object region (Osaka City) is estimated.The authors are developing the ‘stock model’, which es-

timates average value or distribution of appliance’s energye2ciency and dwelling’s thermal properties in the objectregion.

2.2. Schedule of living activities

To determine the occupants’ schedule of living activities,the result of national time use survey by Broadcasting Cul-ture Research Institute, NHK (Japan Broadcasting Corpora-tion) [3] is applied.This data describes the probability distribution of each

living activity, such as sleep, meal, work, etc. at 15 min timesteps on weekdays, Saturday and Sunday for each familymember’s category. The family member’s category is clas-siGed by gender, age and occupation. The authors summedup these data into ‘gender’, ‘age groups’ and ‘employmentstatus’ as shown in Table 1.

2.3. Linking between living activities and energy use

SHASE (The Society of Heating, Air-Conditioning andSanitary Engineers of Japan) developed the simulation soft-ware ‘SCHEDULE’ [4] which generate the schedule of heat-ing and cooling, lighting and other energy use in dwellingsbased on the NHK national time use survey. By referringto this model, each living activity is linked with heatingand cooling, lighting, hot water use and other energy use.

Y. Shimoda et al. / Building and Environment 39 (2004) 959–967 961

Table 1Household types and number of households in Osaka City

Number of Family type Activities of Category Detached house Apartment househousehold of household husband and wifemembers

1 Male 1a 26,166 188,118Female 1b 19,774 142,160Aged male 1c 7221 12,578Aged female 1d 31,400 27,292

2 couple A 2a 21,239 49,088B 19,291 44,584C 2b 3045 7037D 3369 7787

Aged couple 2c 37,600 23,358Mother and C 2d 12,571 19,868child D 2e 8822 13,943Aged mother and child 3606 5699

3 Couple and A 3a 25,676 33,615child B 29,818 39,038

C 1892 2477D 3b 2784 3645

Aged couple and child 3180 4163Mother and C 3c 6156 9729children D 3d 4320 6827Aged mother and children 1766 2791

4 Couple and A 4a 23,478 30,737children B 27,266 35,696

C 1730 2265D 4b 2546 3333

Aged couple and children 2907 3806Couple, child A 4a 5332 1450and parent B 6192 1683

C 4b 393 107D 578 157

5 Couple and A 5a 6287 8231children B 7301 9559

C 463 607D 5b 682 893

Aged couple and children 779 1019Couple, A 5a 3998 1087children and B 4644 1262parent C 5b 295 80

D 434 118

6 Couple and A 6a 846 1108children B 983 1286

C 62 82D 6b 92 120

Aged couple and children 105 137Couple, A 6c 1415 140children and B 1643 163parents C 6d 104 10

D 153 15Couple, A 6c 1431 389children and B 1662 452parent C 6d 105 29

D 155 42

7 ∼ Couple, A 7a 1654 164children and B 1920 190parents C 7b 122 12

D 179 18

Note: Employment status of husband and wife A: Both husband and wife are employed. B: Husband is employed, wife is not employed. C: Husbandis not employed, wife is employed. D: Neither husband nor wife is employed.

962 Y. Shimoda et al. / Building and Environment 39 (2004) 959–967

Table 2Power consumption and standby power of home electric appliances used in this simulation [W]

Appliance Room Power Consumption

ON Stand-by

a. Rice cooker Kitchen 225 31b. Microwave Kitchen 200 —c. Thermos Kitchen — 66d. Fan Kitchen 20 —e. Washing machine Bathroom 126 —f. Tumble dryer Bathroom 1300 —g. Hair dryer Bathroom 450 —h. Desk lamp Bedroom 30 —i. Vacuum Living room 200 —j. Iron Living room 500 —k. Television Living and bedroom 120 2l. VCR Living and bedroom 20 11m. Radio Living and bedroom 100 14n. PC Bedroom 300 1.5o. Stereo Bedroom 100 14p. Kotatsu (foot warmer) Living room 500 —q. Electric carpet Living room 580 —r. Fax Living room — 20s. Telephone Living room — 5t. Refrigerator Kitchen — 60a

u. Shower toilet Toilet — 35

aModeled as a function of outdoor air temperature.

The room where these energy uses occur is also identiGedby the links between each family member and room pro-vided as input data. Home electric appliances are classiGedinto two categories: one is the appliance whose operation islinked with the occupant’s living activities. The other is theappliance whose operation is not linked with living activi-ties such as a fax, a telephone, a refrigerator and a showertoilet. Table 2 shows the power consumption and standbypower of home electric appliances used in this simulation.Approximate dissemination ratios of home electric appli-

ances are also considered in this model. Table 3 shows thedissemination ratio and the average number per householdof home electric appliances used in this study. It is assumedthat TV sets are installed in the living room and all of thebedrooms in order to set the average number per householdto 2.0. However, for the other appliances that average num-ber per household are larger than 1.0, the value is approxi-mated to 1.0.Energy consumption of refrigerators is modeled as a func-

tion of outdoor air temperature.Fig. 2 shows the simulated hourly electricity consumption

of appliances in the family of a couple and two children onweekday.

2.4. Hot water supply

Energy demand for hot water supply is calculated fromthe amount of hot water at each time step, hot water tem-perature and city water temperature. City water temperature

Table 3Dissemination ratios of home electric appliances

Appliance Dissemination Average numberratio (%) per household

Rice cooker 81.2 0.882Microwave 96.4 1.011Washing machine 99.4 1.066Tumble dryer 28.1 0.287Vacuum 97.5 1.489Television 99.5 2.273Radio 48.8 0.571Stereo 78.9 1.420PC 38.1 0.481Kotatsu (foot warmer) 84.8 1.171Electric carpet 73.3 1.102Fax 39.2 0.399VCR 81.0 1.271Telephone 92.6 2.485Refrigerator 99.2 1.231Shower toilet 43.2 0.494

is calculated from outdoor air temperature as follows [5]:

TW = 1:28 + 0:19Ta1 + 0:72Ta2: (1)

TW: City water temperature (◦C);

T a1: Daily average outdoor air temperature

on the same day (◦C);

Y. Shimoda et al. / Building and Environment 39 (2004) 959–967 963

00 2 4 6 8 10 12 14 16 18 20 22

100

200

300

400

500

600

Ele

ctri

c co

nsum

ptio

n [w

/hou

seho

ld]

Others Television Tumble DryerThermos Rice cooker Refrigerator

Fig. 2. Hourly electricity consumption of appliances (a couple and twochildren, weekday).

00 2 4 6 8 10 12 14 16 18 20 22

500

1,000

1,500

2,000

2,500

3,000

3,500

Ene

rgy

for

hot w

ater

sup

ply

[kJ/

h ho

useh

old]

BathWash basinKitchen

Fig. 3. Hourly energy consumption for hot water supply (a couple andtwo children, weekday in winter).

Ta2: Average outdoor air temperature

for past seven days (◦C):

Energy consumption for hot water supply is calculatedfrom dissemination ratio of gas boiler (96.8%) and elec-tric water heater (3.2%) and their assumed e2ciencies (Gasboiler: 0.90, Electric water heater: 0.75). Fig. 3 shows thesimulated hourly energy consumption for hot water supplyin the family of a couple and two children on weekday inwinter.

2.5. Heat load calculation and HVAC energy use

After classifying all dwellings in Osaka City into 20 cate-gories by dwelling type and Joor area and setting represen-tative Joor plan of each category, heating and cooling loadsof these dwellings are simulated.

1-10 0 10 20 30 40

3

5

7

9

11

13

Outdoor air temperature [°C]

CO

P

Base Case coolingTop Runner coolingBase Case heatingTop runner heating

Fig. 4. Relationships between air temperature and COP of room airconditioner.

The algorithm of the simulation is based on HASP/ACLDProgram and the response-factor method is used to calcu-late heat conduction at exterior wall. Ventilation and heatconduction among rooms are not considered in this simula-tion. The degree of thermal insulation is set as the averagethickness of thermal insulation materials in 1997 estimatedby Jukankyo Research Institute [6] for detached house andapartment house, respectively.For heating and cooling by room air conditioner (RAC),

its COP is modeled as the function of outdoor air temper-ature. This function is derived from the measured electric-ity input and heat output for Gve conditions deGned by theJapan Refrigeration and Air-Conditioning Industry Associa-tion (JRAIA) [7]. COP at standard condition is set to the av-erage value of present stock of RAC estimated by JukankyoResearch Institute [6]. Fig. 4 shows the relationships be-tween outdoor air temperature and COP. The amount of en-ergy used for heating by appliances other than RAC such asgas/oil heater is estimated by setting the share and e2cien-cies of these appliances.

2.6. Classi8cation of all families in Osaka City

Using the result of National Population Census 2000(Statistics Bureau and Statistics Center [8]), all householdsin Osaka City (Total: 1128 thousand households) are clas-siGed into 55 household types, 2 dwelling types (detachedhouse or apartment house) and 10 dwelling Joor areagroups. Table 1 also shows the household types and numberof households in Osaka City. In the simulation, householdtypes are summarized into 23 categories as shown in Table1. Table 4 shows the categories and ratio of dwellings be-longing to each category for detached houses and apartmenthouses in Osaka City.Annual energy consumptions are simulated for all 460

categories of household for every 15 min (appliances andhot water supply) or for every 1 hr (cooling and heating).These results are multiplied by household numbers of each

964 Y. Shimoda et al. / Building and Environment 39 (2004) 959–967

Table 4Dwellings Joor area groups and ratio in Osaka City

Scale (m2) Detached house (%) Apartment house (%)

∼ 19 16.7620–29 17.81 16.4330–39 14.2040–49 11.88 12.6350–59 10.05 13.6960–69 11.45 13.9670–79 6.90 7.1580–89 6.58 3.7690–99 8.08100–119 8.37 0.68120–149 8.36 0.74150 ∼ 10.53

1,5000 50 100 150 200

2,000

2,500

3,000

3,500

4,000

4,500

Total Floor Area[m2]

Ann

ual e

lect

rici

ty c

onsu

mpt

ion

[kW

h/ho

useh

old]

1a 1b 1d 2b 2c 3a3b 4a 4b 5b 6d 7b

Fig. 5. Electricity for appliances [detached house](symbols in the legendare correspond to Table 1).

category and aggregated into the annual energy consumptionfor Osaka City’s residential sector.

3. Simulation results

3.1. Simulation results of each household category

Fig. 5 shows the annual electricity consumption by homeelectric appliances for various household types and Joorarea groups in detached houses. For single family, electricityconsumption by home electric appliances is constant withJoor area. On the other hand, electricity consumption forlarge member’s family is decreased with decreasing Joorarea since share of appliances such as watching TV togetheris occur in such conditions.Fig. 6 shows the simulation result for annual electricity

consumption for lighting. It depends strongly on Joor area.In addition, it also depends on household type since numberof occupied room and length of occupied hour are di5erent.Figs. 7 and 8 shows annual energy consumption for heat-

ing and cooling, respectively. Energy consumption for heat-ing and cooling also depends on Joor area. However, in thecase of the households with a small number of members,

00 50 100 150 200

100

200

300

400

500

600

700

800

900

Total floor area [m2]

Ann

ual e

lect

rici

ty c

onsu

mpt

ion

[kW

h/ho

useh

old]

1a 1b 1d 2b 2c 3a

3b 4a 4b 5b 6d 7b

Fig. 6. Electricity for lighting [detached house].

00 50 100 150 200

2,000

4,000

6,000

8,000

10,000

12,000

14,000

16,000

Total Floor Area [m2]

Seco

ndar

y en

ergy

con

sum

ptio

n[M

J/ye

ar h

ouse

hold

]

1a 1b 1d 2b 2c 3a

3b 4a 4b 5b 6d 7b

Fig. 7. Electricity for heating [detached house].

00 50 100 150 200

1,000

2,000

3,000

4,000

5,000

6,000

7,000

Total Floor Area [m2]

Ann

ual E

lect

rici

ty C

onsu

mpt

ion

[kW

h/ye

ar h

ouse

hold

]

1a 1b 1d 2b 2c 3a

3b 4a 4b 5b 6d 7b

Fig. 8. Electricity for cooling [detached house].

energy consumption for heating and cooling no longer in-creases with Joor area for large Joor areas since the numberof rooms heated or cooled does not increase.Figs. 9 and 10 shows the simulation result of total annual

energy consumption per household in detached house andapartment house, respectively. It shows the total energy con-sumption depends on household types strongly than Joor

Y. Shimoda et al. / Building and Environment 39 (2004) 959–967 965

00 50 100 150 200

10,000

20,000

30,000

40,000

50,000

60,000

70,000

Total Floor Area [m2]

Seco

ndar

y en

ergy

con

sum

ptio

n[M

J/ye

ar h

ouse

hold

]

1a 1b 1d 2b 2c 3a

3b 4a 4b 5b 6d 7b

Fig. 9. Total energy consumption [detached house].

00 50 100 150 200

10,000

20,000

30,000

40,000

50,000

60,000

70,000

Total Floor Area [m2]

Seco

ndar

y en

ergy

con

sum

ptio

n [M

J/ye

arho

useh

old]

1a 1b 1d 2b 2c 3a3b 4a 4b 5b 6d 7b

Fig. 10. Total energy consumption [apartment house].

area. In the same category, energy consumption in apart-ment house is smaller than detached house.

3.2. Total energy consumption of residential sector inOsaka City

Total energy consumption of residential sector in OsakaCity that is calculated from the simulation results is shownas the right bar in Fig. 11. The statistic value of energy con-sumption in the FY 1999 and the estimated values derivedby multiplying total number of households in Osaka Citywith unit energy consumption per household estimated fromGeld survey [6] are also shown as the center and right barin Fig. 11.Total energy consumption calculated from the simulation

result is 32; 018 TJ=year, and it is smaller than the statisticvalue by 18%. In addition to the errors in the set valuesand assumptions, the reason for the di5erence is that thesimulation consider only the reasonable energy use and doesnot consider the unreasonable one, e.g., leaving lighting,air-conditioning and appliances switched-on while the roomis unoccupied.

0

10,000

20,000

30,000

40,000

50,000

Unit energy/household* number of households

Statistic value Simulation result

Seco

ndar

y en

ergy

con

sum

ptio

n in

O

saka

City

[T

J/ye

ar]

CoolingHeatingHot waterKitchenLightingAppliances

Fig. 11. Comparison among simulation result, statistic value and estimationfrom Geld survey.

8551,099

3,430

5,700

2,5542,538

2,4382,495

0

1,000

2,000

3,000

4,000

5,000

6,000

No insulation Base case 1992 Standard 1999 Standard

Seco

ndar

y en

ergy

con

sum

ptio

n [T

J/ye

ar c

ity]

Heating Cooling

Fig. 12. E5ect of thermal insulation standard.

On the other hand, the values estimated from the unit en-ergy consumption are larger than the statistic value. Sincethe unit energy consumption per household is usually sur-veyed for household with more than three members, the re-sult tend to overestimate for large cities where single or twopeople families occupies a large portion.In the comparison between the simulation result and es-

timation from the unit energy consumption from the view-point of the composition of use, it is clear that appliancesand heating energy use of the simulation result is smallerand cooling energy consumption is larger than unit energyconsumption method.

3.3. Evaluation of energy saving standard for thermalinsulation and room air conditioner

To examine the e5ectiveness of thermal insulation stan-dards for residential building, energy consumption underthree conditions; no insulation, 1992 standard (‘New EnergyConservation Standard’) and 1999 standard (‘Next Gener-ation Energy Conservation Standard’), are simulated. The1999 standard is stricter than the 1992 standard and thermalinsulation of base case is not adapted to the 1992 standard.Simulation results of annual energy consumption by res-

idential sector in Osaka City are shown in Fig. 12. Heatingenergy consumption is reduced greatly by insulation whilecooling energy consumption does not depend on insulation.If all dwellings in Osaka City are adapted to 1999 standard,

966 Y. Shimoda et al. / Building and Environment 39 (2004) 959–967

0

200

400

600

800

1,000

Heating by RAC Cooling

Ann

ual E

lect

rici

ty C

onsu

mpt

ion

[GW

h/ci

ty]

Base Case Tup runner standard

Fig. 13. Simulated energy consumptions of room air conditioner underthe base case and ‘Top runner standard’ case.

total energy consumption in the residential sector is reducedby 7.7%.As well as thermal insulation standard, Japanese gov-

ernment incorporated new appliance energy standard (com-monly known as ‘Top-Runner Approach’) into the LawConcerning Rational Use of Energy in 1999. In this stan-dard, averaged energy e2ciency in 2004 must be higher thanthe most e2cient model in 1999. For example, the averagevalue of cooling and heating COP of a room air conditionermust be larger than 4.90 for the models between 2.5 and3:2 kW. Therefore energy e2ciency of Japanese applianceshas been improved rapidly in recent years. Fig. 13 showsthe comparison of simulated annual electricity consumptionby room air conditioners in Osaka City. By this standard,energy consumption of residential sector in Osaka City isreduced by 4.0%.The Japan Refrigeration and Air-Conditioning Industry

Association [7] proposed the estimation method of annualpower consumption for room air conditioner and this methodhas been used generally. However, this method has a prob-lem with accuracy since unrealistic conditions (e.g. RAC isoperated 18 hr=day) are assumed. To examine the accuracyof this estimating method, simulated result is compared withthe estimated energy saving by the JRAIA’s method. Thenumber of RACs in Osaka City is set as same as the simu-lation result. The result of estimation by JRAIA method isshown in Fig. 14. It is clear that both annual electricity con-sumption and the amount of conserved energy by JRAIAmethod are overestimated very much.All of these things make it clear that the simulation model

which is developed in this study is essential to estimate thecity- or regional-scale e5ect of various energy conservationpolicies accurately.

3.4. Evaluation on the in:uence of urban heat island

In large Japanese cities, outdoor air temperature isusually higher than the surroundings by urban heat island

0

500

1,000

1,500

2,000

2,500

Heating by RAC Cooling

Ann

ual E

lect

rici

ty C

onsu

mpt

ion

[GW

h/ci

ty]

Base Case Tup runner standard

Fig. 14. Estimated energy consumptions of room air conditioner by JRAIAmethod.

0

200

400

600

800

1000

1200

1400

1600

0:00 8:00 16:00 0:00 8:00 16:00 0:00 8:00 16:00Ele

ctri

city

con

sum

ptio

n in

Osa

ka C

ity

[MW

/Cit

y]

Osaka Hirakata

20

22

24

26

28

30

32

34

36

38

40

0:00 8:00 16:00 0:00 8:00 16:00 0:00 8:00 16:00

Out

door

air

tem

pera

ture

[°C

]

Osaka Hirakata

Fig. 15. Electricity load curves and air temperatures of Osaka weatherdata and Hirakata weather data.

phenomenon. Energy consumption for heating, coolingand hot water supply is supposed to be a5ected by thisphenomenon. To evaluate this e5ect quantitatively, annualenergy consumption using weather data of Hirakata City,which is 25 km to the north-east of Osaka City, is simulated.Fig. 15 shows the estimated electricity load curves in sum-

mer using Osaka weather data and Hirakata weather dataand their outdoor air temperatures. In summer daytime, out-door air temperature in Osaka is higher than that in Hirakata

Y. Shimoda et al. / Building and Environment 39 (2004) 959–967 967

14577 15318

343051342438 1999

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

16,000

18,000

Osaka Hirakata

Ann

ual S

econ

dary

Ene

rgy

Con

sum

ptio

n [T

J/ye

ar c

ity]

Hot water Heating Cooling

Fig. 16. Calculated annual energy consumptions under di5erent weatherdata.

by about 2◦C and electricity demand is also higher by about20%.Fig. 16 shows the comparisons of annual energy con-

sumption for cooling, heating and hot water supply. Whileelectricity consumption for cooling in Osaka City is largerthan that in suburb, energy consumption for heating and hotwater supply in Osaka City is smaller than that in suburbsince heating load is decreased and city water temperatureand COP of room air conditioner (heating mode) are in-creased by air temperature increase. As the result, annualenergy consumption in Osaka City is decreased by the urbanheat island phenomenon. The di5erence in this case corre-sponds to 6% of total energy consumption in the residentialsector in Osaka City.

4. Conclusion

In this paper, the simulation model which predicts city-level energy consumption for the residential sector is de-veloped. The unique characteristics of this model are asfollows:

(1) The distribution of household type and dwelling typein the region is considered.

(2) Since energy use of appliances are linked to living ac-tivity of occupants, the e5ect of life style change onenergy consumption can be examined.

(3) The impact of changes in the e2ciencies of appliancescan be evaluated correctly.

(4) Since heat load calculations are included in this model,the e5ect of thermal insulation and di5erences in climatecan be considered.

Therefore, this model can evaluate the e5ect of variousenergy conservation policies. By adding the scenario of howdwellings and appliances stock will change and how thepopulation and the composition of household will change,the model can predict the change of energy consumption inthe future.

Acknowledgements

This research is supported by Environmental SystemProject of Handai Frontier Research Center (Leader: Prof.Kiichiro Tsuji).

References

[1] Jones PJ, Lannon S, Williams J. Modeling building energy use aturban scale. Proceedings of Seventh International IBPSA Conference,2001, p. 175–80.

[2] Michalik G, Khan ME, Bonwick WJ, Mielczarski W. Structuralmodeling of energy demand in the residential sector 1. Developmentof structural models. Energy 1997;22(10):937–47.

[3] Broadcasting Culture Research Institute, NHK (Japan BroadcastingCorporation). NHK data book 2000 national time use survey, Nationaled., Prefectural ed. Japan Broadcast Publishing, 2001 [in Japanese].

[4] SHASE (The Society of Heating, Air-Conditioning and SanitaryEngineers of Japan). Schedule of living activities and energyconsumption in dwellings. Report of the sub-committee on energysimulation in dwellings, 2000 [in Japanese].

[5] Minako Nabeshima. Techniques of statistical data analysis to o5ersuggestive observation for the facility designer. PhD thesis, OsakaCity University, 1998 [in Japanese].

[6] Jukankyo Research Institute. 1999 Energy Handbook for ResidentialSector, the Energy Conservation Center, Japan, Tokyo, 1999 [inJapanese].

[7] The Japan Refrigeration and Air Conditioning Industry Association:Calculating method of annual power consumption for room airconditioners JRA4046. The Japan Refrigeration and Air ConditioningIndustry Association, Tokyo, 1999 [in Japanese].

[8] Statistics Bureau and Statistics Center: 2000 Population Census ofJapan. Statistics Bureau of the Prime Minister’s o2ce, Tokyo 2000[in Japanese].