Optimization of a building integrated solar thermal system with seasonal storage using TRNSYS

Accepted Manuscript Optimization of a building integrated solar thermal system with seasonal storage using TRNSYS Christodoulos N. Antoniadis, Georgios Martinopoulos PII:

S0960-1481(18)30385-9

DOI:

10.1016/j.renene.2018.03.074

Reference:

RENE 9943

To appear in:

Renewable Energy

Received Date: 16 July 2017 Revised Date:

5 March 2018

Accepted Date: 27 March 2018

Please cite this article as: Antoniadis CN, Martinopoulos G, Optimization of a building integrated solar thermal system with seasonal storage using TRNSYS, Renewable Energy (2018), doi: 10.1016/ j.renene.2018.03.074. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT 1

Optimization of a Building Integrated Solar Thermal System with Seasonal

2

Storage using TRNSYS

3

Christodoulos N. Antoniadis, Georgios Martinopoulos*

RI PT

4

International Hellenic University, School of Science and Technology, 14th km Thessaloniki –

6

Moudania, EL 570 01, Thermi – Thessaloniki, Greece, Tel: +302310807533

7

*Corresponding Author: Georgios Martinopoulos

8

ABSTRACT

M AN U

SC

5

In the current work a building integrated solar thermal system with seasonal storage is

10

optimized with the use of TRNSYS modeling software in order to evaluate different

11

integration options of the solar collector array. The model calculates the space heating

12

needs during the heating period as well as the annual domestic hot-water needs of a typical

13

single-family detached home in the city of Thessaloniki, Greece that has been built according

14

to the latest building code. The contribution of the solar system and the thermal load

15

covered by the auxiliary conventional system are determined, for both space heating and

16

domestic hot water, considering the respective simulation periods. The relative solar

17

fractions as well as the combined annual solar fraction are also calculated. A parametric

18

analysis on the impact of various solar collector areas and types, building integration type, as

19

well as the volume of the seasonal storage tank is also presented in order to optimize system

20

design and obtain a seasonal combined solar fraction of at least 39%.

21

Words: 5358

22

Keywords: Solar thermal energy, solar collectors, TRNSYS, seasonal storage

AC C

EP

TE D

9

1

ACCEPTED MANUSCRIPT 23

1. Introduction The building sector represents one of the biggest energy consumers in the European

25

Union (EU), accounting for more than 40% of the final energy consumption (European

26

Environment Agency, 2010). To address this issue, the EU implemented a series of directives

27

that promote the exploitation of energy alternatives for buildings, used primarily for

28

electricity, heating, cooling and the provision of hot water, that led to Directive 2010/31/EC

29

(European Commission, 2010) which defined minimum rules on the performance of buildings

30

and introduced energy certificates, taking into account the external climatic conditions and

31

defining the Net Zero Energy Building (NZEB). These actions have led to a decrease in the

32

final consumption of the residential sector from 316.7 Mtoe in 2000 to 287.1 Mtoe in 2014

33

(ODYSSEE, 2016). The noticed improvement in energy efficiency was a result of better

34

thermal performance of buildings, more efficient electric appliances (air conditioning) and

35

heating systems (condensing boilers and heat pumps). Still, part of this improvement was

36

counteracted by a growing number of electric appliances, larger homes and the dispersion of

37

central heating, which led to an escalation in the average consumption per dwelling by 0.4%

38

per annum, compensating 60% of the energy-efficiency development reached through

39

technological modernization (European Environment Agency, 2016).

AC C

EP

TE D

M AN U

SC

RI PT

24

40

Of the various renewable-energy systems that can be installed in the building sector in

41

order to cover energy requirements (electrical and thermal loads), solar-energy systems are

42

currently the most widely used, mainly in the form of solar thermal and photovoltaic

43

systems. Especially for the southern countries of the EU, with their high annual solar

44

radiation and temperatures, solar-energy systems are already a viable alternative to fossil

45

energy systems and are expected to become even more efficient and cost-competitive in the

46

future (ECOFYS, 2013). Most EU countries throughout the region enjoy high numbers of new 2

ACCEPTED MANUSCRIPT 47

installations annually both in the form of Domestic Solar Hot Water Systems (DSHWS) and of

48

grid-connected photovoltaic systems, while the biggest potential is expected for renewable

49

combi systems that generate heat for space heating purposes in winter times, cooling

50

through air-conditioning systems in summer and domestic hot water throughout the year. In Greece, solar-energy systems enjoy very high penetration rates both in the form of solar

52

thermal systems as well as in the form of photovoltaic. DSHWS are a mature technology,

53

boasting an installed capacity of 4.4 million m² (3000 MWth) at the end of 2015, placing

54

Greece third in the per capita installed capacity in the EU (ESTIF, 2016). The development of

55

the total installed solar thermal capacity in Greece during the last 16 years is presented in

56

figure 1.

M AN U

SC

RI PT

51

Although solar thermal systems are a mature technology, up to now most systems were

58

not integrated into the buildings but were installed in available areas with the notion of

59

function over form. Nowadays, architectural integration is a major issue for the development

60

and spreading of solar thermal technologies, especially considering the necessity of their use

61

as a step towards net zero energy buildings. Two options are available, façade and roof

62

integration, although still the architectural quality of most existing building integrated solar

63

thermal (BIST) systems is considered low (Probst and Rocker, 2007). On the other hand a

64

number of researchers argue that economic and educational issues should be addressed

65

towards the deeper penetration of the BIST technology into the market, especially for the

66

façade integrated solar thermal systems (Capel et al, 2014).

AC C

EP

TE D

57

67

Environmental evaluation of building integrated flat plate and vacuum tube collectors, by

68

conducting a life cycle assessment comparing the two systems, highlighted the considerably

69

better environmental behavior of the vacuum tube configuration compared to the flat plate

70

one (Lamnatou et al, 2015). 3

ACCEPTED MANUSCRIPT Regarding thermal energy storage, various technologies and configurations have been

72

investigated along with solar thermal systems in order to improve the performance of

73

domestic hot water and space heating applications, like an energy storage unit containing

74

phase change material linked to a solar powered heat pump system used for space heating

75

(Essen, 2000), or the possibility of architectural fusion of thermal energy storage systems in

76

buildings (Navarro et al, 2016).

RI PT

71

However, the main obstacle towards the further diffusion of solar thermal systems for

78

space heating applications is the fact that solar potential is low during the heating period. To

79

overcome this issue long-term energy storage is necessary. The main difference between

80

seasonal and short-term storage systems is located on their size, with the first to be much

81

greater than the second (Brown et al, 1981). More specifically, it has been evaluated that

82

storage capacity per unit of collector area should be 100 – 1000 times greater for seasonal

83

storage than for diurnal storage (Duffie and Beckman, 2013).

TE D

M AN U

SC

77

There are three different mechanisms for energy storage, which can therefore be

85

considered for seasonal storage of solar thermal energy as well. These concepts include

86

sensible heat storage, latent heat storage and chemical reaction/thermo-chemical heat

87

storage (Novo et al, 2010). About residential scale thermal storage applications, and

88

particularly those currently used in practice, most of them store energy in the form of

89

sensible heat, while latent and chemical methods are considered promising but not mature

90

enough yet (Schmidt et al, 2004).

AC C

EP

84

91

The sensible heat storage method concerns the storage of heat as internal energy in the

92

form of the temperature increase of a selected material, generally a liquid or solid, which is

93

used as the storage medium. The specific heat of the medium used, as well as the

94

temperature increase determines the amount of sensible heat that is stored. High specific 4

ACCEPTED MANUSCRIPT heat and density are two major desired characteristics for a storage medium concerning its

96

thermal capacity evaluation. The rate of absorbing or releasing heat is another important

97

requirement, depending each time on the specific heat storage application. In addition, the

98

temperature range over which the system operates can also influence the medium choice.

99

Of course, apart from its thermal capabilities, cost and space availability are additional issues

100

that should be considered, especially for residential scale seasonal storage. For all the above

101

reasons water, rock materials (gravel, pebbles) and soil/ground have been typically used as

102

storage media for solar seasonal storage systems in residential applications (Pinel et al,

103

2011).

M AN U

SC

RI PT

95

A number of researchers have investigated Seasonal Solar Thermal Energy Storage (SSTES)

105

systems through TRNSYS modeling in the past. Terziotti et al (2012) studied such a system,

106

modeled for a 15700 m2 student housing building in Virginia. Flat solar collectors of 1930 m2

107

were used, providing heat to the storage bed, which contained sand as the storage medium.

108

Four different bed sizes were considered, with volumes ranging from 5005 to 9295 m3.

109

Simulations regarding the fifth year of operation, when the system was considered to have

110

reached equilibrium, showed that up to 91% of the total energy needs can be provided by

111

the SSTES system in the best-case scenario.

AC C

EP

TE D

104

112

Sweet et al (2012) modeled a SSTES system for a single-family dwelling in Richmond,

113

Virginia, investigated for different-sized floor areas, ranging from 74 to 223 m2. A theoretical

114

flat plate collector was used for the model, covering 39 to 99 m2 of the total roof area,

115

depending on the varying house floor area. Sand was used as the storage medium in an

116

underground storage bed, in order to provide better insulation conditions. Five-year

117

simulations were run to ensure that the system reached a steady state. Results showed that

5

ACCEPTED MANUSCRIPT 118

the utilization of SSTES could reduce auxiliary heat demand requirements by 64% to 77%

119

compared to models without SSTES. Li et al (2014) simulated the performance of a solar thermal heat pump system combined

121

with seasonal energy storage, modeled for a 2252 m2 six-story dorm building in Beijing.

122

Water was used as the storage medium, both for the 105 m3 seasonal storage tank and the

123

9.6 m3 domestic hot-water tank. The operation of the system was simulated for a single-year

124

period, and the outcomes showed that the monthly energy-saving ratio reached 52% when

125

compared to a traditional space heating system. Regarding the domestic hot water, the

126

monthly average solar fraction reached 68%.

M AN U

SC

RI PT

120

McDowell et al (2008) used TRNSYS software to simulate the performance of the Drake

128

Landing Solar Community (DLSC) project in Alberta, Canada. The DLSC includes a 52-single-

129

family house subdivision, where space and water heating are supplied by a system consisting

130

of 800 flat plate solar collectors, two short-term thermal storage tanks, the Borehole

131

Thermal Energy Storage (BTES) system for seasonal energy storage and the district heating

132

supply system. Totally 144 boreholes connected in series were drilled at a depth of

133

approximately 35 meters to create the BTES. By using fifty years of historical weather data,

134

the simulation model predicted that the system will provide an annual solar fraction of 90%

135

after five years of operation.

AC C

EP

TE D

127

136

Sibbitt et al (2012) monitored the performance of the DLSC Seasonal Solar Thermal Energy

137

Storage system, providing measurements of five years and comparing these results against

138

the TRNSYS simulation predictions for the same period. Because of the gradual charging of

139

the BTES system, the monitored performance was observed to increase every year, resulting

140

in a solar thermal contribution of 97% in the fifth year. Regarding the simulation model, it

6

ACCEPTED MANUSCRIPT 141

was presented that for the 5-year period, solar fraction was expected to increase up to 89%

142

over the period. Lundh et al (2008) simulated the performance of an actual seasonal ground storage

144

system of solar heat, constructed to provide low temperature space heating to a residential

145

area with a 550 MWh total heat demand in Anneberg, Sweden. The storage system included

146

60000 m3 of crystalline rock with 100 boreholes drilled at a depth of 65 m, while flat plate

147

solar collectors were installed with a total area of 2400 m2. The TRNSYS simulation resulted

148

in an average solar fraction of about 70% after five years of operation.

SC

RI PT

143

Antoniadis and Martinopoulos (2017) used TRNSYS to model a water-based SSTES system,

150

applied for a 120 m2 flat rooftop detached house in the city of Thessaloniki, Greece. Flat

151

plate solar collectors of 30 m2 and a stratified seasonal storage water tank of 36 m3 were

152

considered. The operation of the designed system was examined for two consecutive years,

153

in order for the storage tank to be fully charged for the heating period. During the second

154

year of the simulation, it was found that the system could cover more than the half of the

155

space heating load requirement.

TE D

M AN U

149

Although a number of researchers have dealt with SSTES, there has been limited work on

157

optimizing these systems while on the same time trying to find the optimum building

158

integration strategies for the solar collector array. To that end, in the present work, TRNSYS

159

modeling software is used to simulate and optimize a building integrated solar thermal

160

system with seasonal thermal energy storage (STES).

AC C

EP

156

161

The model calculates the space heating and domestic hot water needs of a typical 120 m²

162

single-family detached home in the city of Thessaloniki, Greece that has been built according

163

to the latest building code. The contribution of the solar system, as well as the thermal load

164

covered by the auxiliary conventional system is determined, and the seasonal solar fraction 7

ACCEPTED MANUSCRIPT is calculated. A parametric analysis of the impact of various solar collector areas and types,

166

building integration type, as well as the volume of the seasonal thermal energy storage is

167

also presented in order to optimize system design. The combined annual solar fraction in all

168

cases ranges from 39 and up to 67%.

169

2. Building Topology and Solar System Description

RI PT

165

The residential sector was responsible for 29.4% of the total final energy consumption in

171

Greece in 2012. According to a recent survey, every household throughout the country

172

consumes, on average, 14 MWh in order to cover its needs. Thermal needs (for space

173

heating and hot-water production) account for 73%, while the remaining 27% is needed for

174

the various electrical appliances (Hellenic Statistics Authority, 2013). With the incorporation

175

of the Directive 2010/31/EC in the Greek Regulatory Framework, energy consumption for

176

heating is predicted to decline from an average of more than 100 kWh/m²,yr to as low as 15

177

kWh/m²,yr (Tsalikis and Martinopoulos, 2015).

TE D

M AN U

SC

170

As more than 86% of all residences in Greece have a surface area of up to 120m² (Hellenic

179

Statistics Authority, 2014), a south oriented 120 m² detached house with an inclined tiled

180

rooftop, covering the needs of a four-member family, was selected in order to estimate the

181

solar potential from a SSTES system. The building has a rectangular shape with 12 m width

182

and 10 m length, a flat terrace and an internal height of 3 m as presented in Figure 2.

AC C

EP

178

183

The systems under investigation have been designed and simulated in TRNSYS for

184

implementation at a single-family detached home in Thessaloniki, Greece. Thessaloniki is in

185

Northern Greece (40.30°Ν, 22.58°Ε) and its climatic conditions can be marked as atypical

186

Mediterranean with relatively harsh winters and hot, humid summers. Specifically, it is

187

calculated that the Heating Degree Days for Thessaloniki are 1713, at a base temperature of 8

ACCEPTED MANUSCRIPT 188

18°C (Papakostas et al, 2005). Moreover, the average annual temperature is 16.7°C

189

(Papakostas et al, 2014), while the average annual global horizontal irradiation is 1466

190

kWh/m2 (YPEKA, 2010).

192

For the simulations, a weather data file for Thessaloniki city is imported as an external file in Type 109 data reader in the standard TMY2 format.

RI PT

191

For the simulation of the building’s thermal load, TRNSYS offers a number of different

194

options with the most common being Type 56, a complex multi-zone model and Type 12a a

195

simple single zone degree day model (Persson et al, 2011). Although Type 56 is used by most

196

researchers nowadays in building simulations (Razavi et al, 2018), Type 12a is still used in the

197

analysis of renewable energy sources utilization in buildings (Davidsson et al, 2012), (Li et al,

198

2014) and in the optimization of solar heating and cooling systems for building applications

199

(Calise et al, 2010), (Xu and Wang, 2017) among others. The reason is that Type 12a provides

200

a very good approximation of the building loads, especially during the heating period, while

201

at the same time requiring less calculation time, which is helpful when the scope of a study is

202

the optimization of a system (Persson et al, 2011), as in this case.

EP

TE D

M AN U

SC

193

For the above reasons, the dwelling considered, was modelled initially with both models,

204

Type 12a and Type 56. The overall conductance for heat loss from the house (UA value) is

205

defined at 200 W/K according to the calculations provided in table 1 for use in Type 12a. The

206

table presents the structural elements within the building, providing their heat transfer

207

coefficient and the area of each specific element that is exposed to the external air or

208

ground, contributing to the heat flow, that were used in Type 56. According to the technical

209

guidelines, the absorptivity of the opaque outer surfaces is 0.4 for the vertical building

AC C

203

9

ACCEPTED MANUSCRIPT 210

elements and flat roofs, 0.6 for inclined roofs, while the emissivity of the outer surfaces is

211

0.8 (YPEKA, 2010). Apart from the specific and the overall UA values, the average heat transfer coefficient

213

(Um) is also calculated. At this point, it should be noted that the calculations do not consider

214

the integration of the solar collectors into the building envelope, as the thermal

215

transmittance and the solar heat gain coefficient can hardly be determined for building

216

integrated solar thermal elements, while the neglect of the integration usually overestimates

217

the heating demand of the building (Maurer et al, 2017). Therefore, it is considered that the

218

integration of solar collectors in all cases, either in the façade or into the roof, would further

219

decrease the heat-transfer coefficients of the specific elements, contributing to an even

220

lower overall conductance for heat losses. The development of new simulation models,

221

including the relevant coupling of the building integrated solar thermal elements to the

222

envelope, could lead to more precise calculations of the energy flux through such building

223

elements (Maurer et al, 2017).

TE D

M AN U

SC

RI PT

212

The temperature at which the house is to be maintained during the heating season is set

225

at 20°C (YPEKA, 2010). To account for the internal gains, the Type 14c component has been

226

used according to the occupancy scenarios and the appliances/lighting used.

AC C

EP

224

227

A water based STES tank is selected, as water is considered a suitable storage medium

228

solution, with high heat exchange rates, high thermal capacity and low cost, while the

229

construction implementation is rather simple and conventional. Despite its limited

230

temperature range, it is more than sufficient for residential applications operating in the

231

range of 40-70 °C, such as space heating and domestic hot water production. Moreover, its

10

ACCEPTED MANUSCRIPT 232

upper temperature limit is close to the corresponding limit of the fluid exiting typical solar

233

collectors (Pinel et al, 2011). The tank has an orthogonal shape and is equally stratified into five nodes, since this

235

degree of stratification is considered sufficient for annual predictions of the performance of

236

solar storage tanks using TRNSYS simulation (Oberndorfer et al, 1999). Each node has been

237

set at a 0.6 m height. The secondary storage tank for the hot water for sanitary uses

238

performs the same degree of stratification and has a volume of 0.3 m3. As a heating system,

239

low temperature radiators are used, which is the usual choice for this type of systems

240

(Tsalikis and Martinopoulos, 2015).

M AN U

SC

RI PT

234

Regarding the construction materials, the seasonal storage tank is considered to be

242

constructed of reinforced concrete. A polyurethane coating system featuring a stainless steel

243

leafing pigment is placed at the top and the vertical inside parts of the tank providing extra

244

insulation. Considering these specifications, the average tank loss coefficient is set to 1.2

245

kJ/(hr m2 K). Two different options of solar collector types are investigated, both flat plate

246

and vacuum tube collectors in various array sizes and inclinations.

EP

TE D

241

The model of the specific system, created in TRNSYS software through the graphical user

248

interface ‘Simulation Studio’, is presented in Figure 3 while the schematic representation of

249

the solar thermal seasonal storage system is presented in Figure 4. Various components,

250

including inputs, parameters and outputs, represent different individual systems. These

251

components are linked together by connecting the outputs of the one to the input(s) of the

252

other(s), modelling the required system. The total system can be separated into four

253

different subsystems, namely the:

254

• Solar collector subsystem

AC C

247

11

ACCEPTED MANUSCRIPT • Domestic hot-water tank subsystem

256

• Seasonal Thermal Energy Storage tank subsystem

257

• Building heating load subsystem

258

The solar collector subsystem consists of the flat plate collector model (Type 1b) or the

259

evacuated tube collector model (Type 71), as well as the other components that participate

260

in the circuits with the domestic hot water and seasonal storage tanks, depending on the

261

case study. Technical data of commercial collectors available in the Greek market were used

262

for both solar thermal collector types. The required specifications are provided by the

263

manufacturers and resulted from standard tests of collector efficiency versus the ratio ΔT/IT

264

(fluid temperature minus ambient temperature to solar radiation). Moreover, incidence

265

angle modifier available data were used to account for the effects of non-normal solar

266

incidence. Table 2 presents the technical specifications of the solar collectors, including the

267

efficiency coefficients that are used as parameters for the TRNSYS deck file.

TE D

M AN U

SC

RI PT

255

The collector receives hourly meteorological data as inputs from Type 109 data reader in

269

the standard TMY2 format and charges the two tanks according to the user defined control

270

specifications. The control pattern has been set up in such a way that only one of the storage

271

tanks is provided with thermal energy by the solar collector (if there is enough solar

272

radiation) for each time step, placing priority to the DHW tank.

AC C

EP

268

273

A differential controller (Type 2) is utilized for each collector-tank circuit, receiving inputs

274

of the fluid temperature that exits the collector Thigh and the temperature of the fluid at the

275

bottom node of the storage tank Tlow. The temperature difference between them is

276

compared to two predefined dead band temperature differences, generating an output

277

control function which may be ON (1) or OFF (0). In case that the last controller status is off,

278

the temperature difference Thigh - Tlow is compared to the upper dead band temperature 12

ACCEPTED MANUSCRIPT difference ΔΤΗ, giving an output control function with value 1 when Thigh - Tlow ≥ ΔΤΗ. The

280

value of the output control function will remain 1 and the controller status on, until the

281

temperature difference Thigh - Tlow becomes lower than the lower dead band temperature

282

difference ΔΤL. Regardless of these conditions, the controller constantly monitors the top

283

node temperature of the tank and sets the control function to zero if this temperature

284

exceeds the predefined high limit cut-out value. The output control functions are linked to

285

the pumps, thus switching them on or off, permitting or not the charging of the tanks.

RI PT

279

In order to ensure just a single tank charging per time step, while providing priority to the

287

DHW tank, the output control signal of the DHW circuit differential controller (0 or 1) is

288

linked to the input control function (γ) of the flow diverter component (Type 11f), setting the

289

relevant position of a damper that drives the entire flow rate to one of the two exits towards

290

one of the two tanks. Specifically, when the value of the output control signal of the DHW

291

differential controller is 1, then the input control function (γ) is 1 and the flow rate of the

292

collector is directed to the DHW tank. In another case, when the DHW circuit’s controller

293

output signal is 0, the fluid exiting the solar collector is driven to the STES tank, which is

294

charged if the pump of that circuit is switched on. At any case, not more than one storage

295

tank could be provided with solar thermal energy by the solar collector for each time step.

296

Since two different outputs cannot be linked to the same input, a tee-piece (Type 11h) is

297

used to close the fluid loops with the solar collector, although the two inlet liquid streams

298

are never mixed.

AC C

EP

TE D

M AN U

SC

286

299

The DHW tank subsystem includes the relevant stratified storage tank (Type 4c), the

300

components placed to define the demand for DHW, as well as the components used to

301

simulate the contribution of colder fluid that is supplied when the tank flow stream to the

302

load exceeds the required temperature Tset. The last operation is achieved by the use of a 13

ACCEPTED MANUSCRIPT temperature-controlled flow diverter (Tempering Valve - Type 11b). In this case, a control

304

function is internally calculated so that an appropriate flow stream displaces fluid of

305

temperature Th (tank top node temperature), so that the mixed fluid temperature of the

306

outlet of the tee-piece 2 will not exceed the required set point temperature Tset. When the

307

thermal energy from the solar collector is inadequate for achieving the minimum required

308

tank temperature, an auxiliary heating element is activated in the tank.

RI PT

303

Regarding the DHW draw profile, Type 14 components were used to simulate the

310

everyday demand for DHW, by creating a water draw time dependent forcing function.

311

Based on the daily DHW consumption of 50 liters per person, that is proposed by the Greek

312

legislation scheme (KENAK) for residential buildings, a daily demand of 200 liters was

313

distributed throughout a 24h cycle, according to daily use patterns that had been

314

investigated for Greece (Papakostas et.al, 1995). Figure 5 depicts the flow rate of the DHW

315

according to the time-dependent daily needs, as well as the cumulative demand throughout

316

the day. It has to be noted that the volume of 50 lt/day can be considered at the upper limit,

317

as evidenced by relevant studies (Papakostas et.al, 1995), thus the actual fraction covered

318

can be even higher.

EP

TE D

M AN U

SC

309

The STES tank subsystem includes a large stratified storage tank (Type 4c) modeling the

320

long-term storage of solar thermal energy during the charging period for heating application

321

during the heating period. Since the storage tank is assumed to be buried under the ground,

322

the environment temperature input is linked with the output of the ground temperature

323

model (Type 501). This component is used to predict the soil temperature in a specific depth

324

where the STES tank is considered to be placed.

AC C

319

325

The building heating load subsystem has been designed to model a building which space

326

heating load should be covered during the heating period. The heating load is partially 14

ACCEPTED MANUSCRIPT covered by the STES tank, as well as by an auxiliary heating source with parallel operation

328

when the tank is not able to cover the entire load by itself. Type 12a component, modeling a

329

single zone structure using the energy/ (degree-day) concept and the Type 56 component,

330

have both been used initially for this case. The user defined specific parameters of the

331

building, such as the overall conductance, as well as meteorological TMY2 input data from

332

Type 109 reader are utilized to calculate the demand for heating and the contribution from

333

each source (STES tank/auxiliary) to meet that demand. The house thermal losses and the

334

possible gains (solar and internal) are calculated at each time step, defining the amount of

335

energy required by the house to maintain the desired set point temperature during the

336

entire heating season.

SC

M AN U

337

RI PT

327

3. Simulation Results and Discussion

The performance of the system is investigated concerning the space heating contribution

339

during the heating period as well as the DHW load coverage during the entire season. With

340

regard to the heating period, it is always defined as a specific time period during the year,

341

when the heating load is on demand. The estimation of this period usually differs from

342

country to country according to the existing meteorological conditions. Actually, it also may

343

be diversified for different regions inside a country. The heating period in the city of

344

Thessaloniki, which belongs to the third climatic zone of Greece based on the degree-days

345

for heating (YPEKA, 2010), is assumed as the period from October 18th to April 23rd.

346

Therefore, this specific period has been set at the TRNSYS control cards to account for the

347

space heating demand that is calculated at 6354 kWh with Type 12a and at 6558 kWh with

348

Type 56 which is a difference of less than 3.2%. The difference between the results from the

349

two models is consistent with previous studies regarding their comparison during the

AC C

EP

TE D

338

15

ACCEPTED MANUSCRIPT heating period (Persson et al, 2011). As the difference between the two components used

351

for the calculation of the buildings heating loads are very close, for all the case studies Type

352

12a was used as the best compromise between accuracy and speed. Regarding the DWH

353

demand, it is estimated at 2552 kWh annually, contributing to a total heating demand (space

354

heating and hot-water demand) of 8906 kWh throughout the year.

RI PT

350

The operation of the system has been examined for two consecutive calendar years in

356

order to achieve steady-state conditions. Prior studies have proven that the different initial

357

conditions, concerning the charging or not of the stratified water tanks during the preceding

358

cooling period, diversify the simulation parameters of the two years and the final results as

359

well (Antoniadis and Martinopoulos, 2017).

M AN U

SC

355

In order to optimize the system, two distinct type of solar collectors and two different

361

building integration options (which in turn influence the collector array size) are investigated.

362

Initially, a roof integrated flat plate solar array (with an inclination of 35°) was assumed with

363

an installed area ranging from 14 to 30 m² (in increments of 4 m²). As a second option, a

364

vacuum tube solar array was also investigated, either integrated in the façade (inclination of

365

90°) with an area of 14 to 20 m², or integrated into the roof with an installed area of 22 to 30

366

m² (in increments of 2 m²). Furthermore, the impact from the size of the orthogonal STES

367

tank was also investigated. To that end, a tank with volume ranging from 5 to 20 m3 (in

368

increments of 5 m3) was combined with the different solar collector arrays. The tank in all

369

cases has a height of 3m and a width of 1.5m and the only dimension that changes is the

370

length of the tank.

AC C

EP

TE D

360

371

In figures 6 and 7, the annual coverage for all the investigated options is presented. It is

372

apparent that the impact of the solar collector array is more pronounced than that of the

373

storage tank volume, regardless of the solar collector type. The results also highlight the 16

ACCEPTED MANUSCRIPT 374

versatility of vacuum collectors, as their integration on the façade of the building does not

375

impact the solar fraction, as they are able to cover the same heating needs as a flat plate

376

inclined solar collector roof array with the same total area. Taking into consideration the energy demands, the comparative cost of the scenarios

378

analyzed as well as building integration needs, the use of a 20m² vacuum tube array coupled

379

with a 5 m3 STES tank is the optimum solution, as it can cover at least 44% of the total

380

heating loads, while at the same time leaving the roof available for possible integration of

381

photovoltaic in order to cover the electricity needs of the building. The use of façade

382

integrated collectors has the added benefit of a better U value than inclined collectors,

383

because heat losses from the collector are reduced as a result of lower convection levels

384

between the absorber and the glazing, while also the overall U value of the wall façade

385

decreases as well (Weiss, 2003).

M AN U

SC

RI PT

377

Under the conditions stated above, the graphic plots of the simulations obtained from the

387

online plotter 3 in TRNSYS, examining the scenario that provides the highest annual solar

388

coverage (roof integrated 30 m² vacuum tube collectors at an inclination of 35° and a STES

389

with a 20 m3 volume) for the second year’s heating period of October 18th - April 23rd, are

390

displayed in figures 8 and 9. Figure 8 presents the instantaneous space heating load (kW) as

391

well as the aggregate space heating demand (kWh) for the whole heating period, while figure

392

9 displays the contribution of the STES and the auxiliary heater during the same period.

AC C

EP

TE D

386

393

As it can be seen from figure 9, the STES tank, has already reached its maximum

394

temperature at the end of the charging period and is able to meet the entire space heating

395

load of the building for more than two consecutive months, from the beginning of the

396

heating period on October 18th until December 23rd, covering 1864 kWh of heating

17

ACCEPTED MANUSCRIPT 397

requirement. At the end of the heating period, the contribution of the STES tank (3819 kWh)

398

accounts for 60% of the total demand. Table 3 presents the temperature values of each of the five tank nodes on October 18th,

400

which have been set equal to that obtained at the end of the first year’s charging period

401

simulation, as well as the temperatures at the end of the heating period on April 23rd.

RI PT

399

The thermal performance of the STES tank during the charging and the heating period for

403

the best case scenario is presented in figures 10 and 11. Figure 10 displays the rate of

404

thermal energy transfer from the collector to the storage tank, the rate of thermal energy

405

loss from the tank to the environment (negative values), as well as the corresponding

406

aggregate energy quantities during the first year’s charging period (April 24th – October 17th),

407

when no space heating load is demanded. In figure 11 the heat transfer rates during the

408

second year’s heating period (October 18th – April 23rd) are presented, displaying the rate at

409

which thermal energy is removed from the tank to supply the space heating load (negative

410

values) in addition to the charging and heat loss rates. Moreover, the aggregate charging and

411

discharging energy quantities during the same period are also presented.

EP

TE D

M AN U

SC

402

4. Conclusions

413

The performance of a building integrated solar thermal system, which utilizes seasonal

414

thermal energy storage for domestic applications, has been investigated in order to optimize

415

its sizing. The system could cover almost 67% of the total heating load requirement for a

416

typical building in Thessaloniki, Greece while also minimizing the visual impact, while in all

417

cases (56 in total) the proposed system could cover at least 37% of the total needs.

418

However, due to the increased heat demand and the extended tank discharge during the

AC C

412

18

ACCEPTED MANUSCRIPT winter months, the operation of an auxiliary source is necessary, in order to meet the entire

420

load.

421

References

422

Antoniadis, C.N., Martinopoulos, G. (2017) “Simulation of Solar Thermal Systems with

423

Seasonal Storage Operation for Residential Scale Applications”, Procedia Environmental

424

Sciences, 38, 405 – 412.

425

Braun, J. E., Klein, S. A., & Mitchell, J. W. (1981). “Seasonal storage of energy in solar

426

heating.” Solar Energy, 26(5), 403-411.

427

Calise, F., d’Accadia, M. D., & Palombo, A. (2010). “Transient analysis and energy

428

optimization of solar heating and cooling systems in various configurations.” Solar Energy,

429

84(3), 432-449.

430

Cappel, C., Streicher, W., Lichtblau, F., Maurer, C. (2014) “Barriers to the market penetration

431

of façade-integrated solar thermal systems”, Energy Procedia, 48, 1336 – 1344.

432

Davidsson, H., Perers, B., & Karlsson, B. (2012). “System analysis of a multifunctional PV/T

433

hybrid solar window.” Solar Energy, 86(3), 903-910.

434

Duffie, J.A., Beckman, W.A. (2013) “Solar Engineering of Thermal Processes”, John Wiley &

435

Sons, New York.

436

Ecofys, “Towards nearly zero-energy buildings - Definition of common principles under the

437

EPBD -Final report”, 2013.

438

Esen, M. (2000) “Thermal performance of a solar-aided latent heat store used for space

439

heating by a heat pump”, Solar Energy, 69(1), 15-25.

440

European Commission, Directive 2010/31/EC, “On the Energy Performance of Buildings”,

441

2010.

AC C

EP

TE D

M AN U

SC

RI PT

419

19

ACCEPTED MANUSCRIPT 442

EEA, “Consumption and the Environment – State and Outlook”, 2010.

443

European Environment Agency, “Energy efficiency and energy consumption in the household

444

sector”,

445

efficiency-and-energy-consumption-5/assessment (Accessed: 05 December 2016)

446

European Solar Thermal Industry Federation, “Solar Thermal Markets in Europe 2014”, 2015.

447

Hellenic Statistics Authority, “Survey on Energy Consumption of Households 2011-12”, 2013.

448

Hellenic Statistics Authority, “Building Stock Data from the 2011 Census”, 2014.

449

Lamnatou, Chr., Cristofari, C., Chemisana, D., Canaletti, J.L. (2016) “Building-integrated solar

450

thermal systems based on vacuum-tube technology: Critical factors focusing on life-cycle

451

environmental profile”, Renewable and Sustainable Energy Reviews, 65, 1199–1215

452

Li, H., Sun, L., Zhang, Y. (2014) “Performance investigation of a combined solar thermal heat

453

pump heating system”, Applied Thermal Engineering, 71, 460-468.

454

Lundh, M., Dalenback, J.-O. (2008) “Swedish solar heated residential area with seasonal

455

storage in rock: Initial evaluation”, Renewable Energy, 33, 703–711.

456

Maurer, C., Cappel, C., Kuhn, T.E. (2017) “Progress in building-integrated solar thermal

457

systems”, Solar Energy, 154, 158-186.

458

Mc Dowell, T.P., Thornton, J.W. (2008) “Simulation and model calibration of a large-scale

459

solar seasonal storage system”, Third National Conference of IBPSA-USA

460

Navarro, L., Gracia, A., Colclough, S., Browne, M., McCormack, S.J., Griffiths, P., Cabeza, L.F.

461

(2016) “Thermal energy storage in building integrated thermal systems: A review. Part 1.

462

active storage systems”, Renewable Energy, 88, 526-547

463

Novo, A.V., Bayon, J.R., Castro-Fresno, D. and Rodriguez-Hernandez, J. (2010) “Review of

464

seasonal heat storage in large basins: Water tanks and gravel–water pits”, Applied Energy,

465

87(2), pp. 390–397

at:

http://www.eea.europa.eu/data-and-maps/indicators/energy-

AC C

EP

TE D

M AN U

SC

RI PT

Available

20

ACCEPTED MANUSCRIPT Oberndorfer, G., Beckman, W.A., Klein, S.A., “Sensitivity of annual solar fraction of solar

467

space and water heating systems to tank and collector heat exchanger model parameters”,

468

Solar 99 conference: Proceedings of ASES annual conference, 1999.

469

ODYSSEE, Energy efficiency trends & policies (2001) Available at: http://www.odyssee-

470

mure.eu/ (Accessed: 06 December 2016)

471

Papakostas, K., Kyriakis, N. (2005) “Heating and cooling degree-hours for Athens and

472

Thessaloniki, Greece”, Renewable Energy, 30, 1873–1880.

473

Papakostas, K.T., Zagana-Papavasileiou, P., Mavromatis, T. “Analysis of 3 decades

474

temperature data for Athens and Thessaloniki, Greece – Impact of temperature changes on

475

energy consumption for heating and cooling of buildings”, International Conference

476

ADAPTtoCLIMATE, Nicosia - Cyprus, 2014

477

Papakostas, K.T., Papageorgiou N.E., Sotiropoulos B.A. (1995) “Residential hot water use

478

patterns in Greece”, Solar Energy, 54, pp. 369-374.

479

Persson, H., Perers, B., & Carlsson, B. (2011). “Type12 and Type56: a load structure

480

comparison in TRNSYS.” In World Renewable Energy Congress-Sweden, Sweden (No. 57, pp.

481

3789-3796), Linköping University Electronic Press.

482

Pinel, P., Cruickshank, C.A., Beausoleil-Morrison, I., Wills, A. (2011) “A review of available

483

methods for seasonal storage of solar thermal energy in residential applications”, Renewable

484

and Sustainable Energy Reviews, 15, 3341– 3359.

485

Probst, M. and Christian Roecker. (2007) "Towards an improved architectural quality of

486

building integrated solar thermal systems (BIST)." Solar Energy 81.9 1104-1116.

487

Schmidt, T., Mangold, D. and Müller-Steinhagen, H. (2004) “Central solar heating plants with

488

seasonal storage in Germany.” Solar Energy, 76(1-3), pp. 165–174.

AC C

EP

TE D

M AN U

SC

RI PT

466

21

ACCEPTED MANUSCRIPT Sibbitt, B., McClenahan, D., Djebbar, R., Thornton, J., Wong, B., Carriere, J., Kokko, J. (2012)

490

“The performance of a high solar fraction seasonal storage district heating system – five

491

years of operation”, Energy Procedia, 30, 856 – 865.

492

Stritih, U., & Butala, V. (2004). “Optimization of a thermal storage unit combined with a

493

biomass boiler for heating buildings.” Renewable Energy, 29(12), 2011-2022.

494

Sweet, M. L., & McLeskey, J. T. (2012). “Numerical simulation of underground Seasonal Solar

495

Thermal Energy Storage (SSTES) for a single family dwelling using TRNSYS.” Solar Energy,

496

86(1), 289-300.

497

Terziotti, L. T., Sweet, M. L., & McLeskey, J. T. (2012). “Modeling seasonal solar thermal

498

energy storage in a large urban residential building using TRNSYS 16.” Energy and buildings,

499

45, 28-31.

500

Tsalikis, G. and Martinopoulos, G. (2015) “Solar energy systems potential for nearly net zero

501

energy residential buildings”, Solar Energy, 115, pp. 743–756.

502

Weiss, Werner W. “Solar heating systems for houses: a design handbook for solar

503

combisystems”. Earthscan, 2003.

504

YPEKA, EK407/B/9.4.2010, “Regulation on Energy Performance in the Building Sector –

505

KENAK” (in Greek), 2010.

506

YPEKA, Τ.Ο.Τ.Ε.Ε. 20701-3/2010, “Climatic data of Greek areas” (in Greek), 2010.

507

Xu, Z. Y., & Wang, R. Z. (2017). “Simulation of solar cooling system based on variable effect

508

LiBr-water absorption chiller.” Renewable Energy, 113, 907-914.

AC C

EP

TE D

M AN U

SC

RI PT

489

509

22

ACCEPTED MANUSCRIPT List of Figures:

511

Figure 1: Solar Thermal installed capacity (glazed collectors) in Greece

512

Figure 2: Plan view of the building

513

Figure 3: TRNSYS assembly of the simulation system

514

Figure 4: Schematic diagram of the seasonal solar thermal energy storage system for space

515

heating

516

Figure 5: DWH draw profile

517

Figure 6: Space heating demand during the 2nd year for the best case scenario

518

Figure 7: STES tank and auxiliary source contribution during the 2nd year for the best case

519

scenario

520

Figure 8: STES tank heat transfer processes during the 1st year’s charging period for the best

521

case scenario

522

Figure 9: STES tank heat transfer processes during the 2nd year’s heating period for the best

523

case scenario

524

Figure 10: Annual Solar Coverage for roof integrated flat plate collector options

525

Figure 11: Annual Solar Coverage for façade (14-20m²) and roof integrated (22-30m²)

526

vacuum tube collector options

AC C

EP

TE D

M AN U

SC

RI PT

510

527

23

SC

RI PT

ACCEPTED MANUSCRIPT

529

M AN U

528

Figure 1: Solar Thermal installed capacity (glazed collectors) in Greece.

AC C

EP

TE D

530

24

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

531

EP

TE D

Figure 2: Plan view of the building

AC C

532

25

SC

RI PT

ACCEPTED MANUSCRIPT

EP

TE D

Figure 3: TRNSYS assembly of the simulation system

AC C

534

M AN U

533

26

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 4: Schematic diagram of the seasonal solar thermal energy storage system for space

537

heating

AC C

EP

TE D

535 536

27

RI PT

ACCEPTED MANUSCRIPT

SC

538

Figure 5: DWH draw profile

M AN U

539

AC C

EP

TE D

540

28

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

541

EP

TE D

Figure 6: Annual Solar Coverage for roof integrated flat plate collector options

AC C

542

29

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

543

Figure 7: Annual Solar Coverage for façade (14-20m²) and roof integrated (22-30m²) vacuum

545

tube collector options

AC C

EP

TE D

544

30

SC

RI PT

ACCEPTED MANUSCRIPT

546

EP

TE D

M AN U

Figure 8: Space heating demand during the 2nd year for the best case scenario

AC C

547

31

SC

RI PT

ACCEPTED MANUSCRIPT

548

Figure 9: STES tank and auxiliary source contribution during the 2nd year for the best case scenario

M AN U

549 550

AC C

EP

TE D

551

32

SC

RI PT

ACCEPTED MANUSCRIPT

552

Figure 10: STES tank heat transfer processes during the 1st year’s charging period for the best case scenario

M AN U

553 554

AC C

EP

TE D

555

33

SC

RI PT

ACCEPTED MANUSCRIPT

556

Figure 11: STES tank heat transfer processes during the 2nd year’s heating period for the best

M AN U

557

case scenario

AC C

EP

TE D

558

34

ACCEPTED MANUSCRIPT List of Tables:

560

Table 1: Structural specifications of the building elements

561

Table 2: Technical specifications of the solar collectors

562

Table 3: Temperatures of the STES tank nodes at the beginning and the end of the 2nd year ‘s

563

heating season for the best-case scenario.

AC C

EP

TE D

M AN U

SC

RI PT

559

35

ACCEPTED MANUSCRIPT Table 1: Structural specifications of the building elements Azimuth angle 180 180 180 180 0 0 0 0 90 90 90 90 90 270 270 270 270 -

Ui (W/m2 K) 0.35 0.4 2 2 0.35 0.4 2.2 2.2 0.35 0.4 3.5 2.2 2.2 0.35 0.4 2.2 2 0.3 0.6

Ai (m2) 19.96 9 3.52 3.52 25.07 9 0.25 1.68 16.44 8 2.20 1.68 1.68 16.8 8 1.68 3.52 146 120 398 Um= 0.5 W/m2 K

Ui Ai (W/K) 6.99 3.60 7.04 7.04 8.77 3.60 0.53 3.53 5.75 3.20 7.70 3.53 3.53 5.88 3.20 3.53 7.04 43.80 72.00 200.25

EP AC C

565

External Wall Concrete Wall French Door French Door External Wall Concrete Wall Window Window External Wall Concrete Wall Front Door Window Window External Wall Concrete Wall Window French Door Rooftop Floor

TE D

1 S 2 S 3 S 4 S 5 N 6 N 7 N 8 N 9 W 10 W 11 W 12 W 13 W 14 E 15 E 16 E 17 E 18 19 Sum Average heat transfer coefficient

Element

RI PT

Orientation

SC

a/a

M AN U

564

36

ACCEPTED MANUSCRIPT Table 2: Technical specifications of the solar collectors

Gross Area (m2)

Flat Plate

Evacuated

Collector

Tube Collector

2

4

0.95

Fluid specific heat (kJ/kg K)

4.19

Zero loss efficiency (η0)

0.763

First order coefficient a1 (W/m2K)

3.178

Second order coefficient a2 (W/m2K2)

0.015

Tested flow rate (kg/sm2) Incidence angle modifier constant (b0)

4.19

0.46

0.802

0.005

0.02

0.02

0.15

-

AC C

EP

TE D

567

0.93

M AN U

Absorptivity

RI PT

Technical Specifications

SC

566

37

ACCEPTED MANUSCRIPT 568

Table 3: Temperatures of the STES tank nodes at the beginning and the end of the 2nd year ‘s

569

heating season for the best-case scenario. April 23rd

Node 1

95.0°C

40.0°C

Node 2

93.0°C

Node 3

92.0°C

Node 4

90.0°C

Node 5

87.0°C

RI PT

October 18th

38.0°C 37.0°C

36.5°C

SC

Temperature at STES tank

AC C

EP

TE D

M AN U

570

35.0°C

38

ACCEPTED MANUSCRIPT

Highlights: • TRNSYS is used in order to optimize a Seasonal Solar Thermal Storage System • Fifty six different case studies have been simulated Roof and façade integration options for the solar collectors were investigated

RI PT

•

AC C

EP

TE D

M AN U

SC

• Solar fraction ranged from 37 to 68% of the total thermal load