21 - 26 April 2018 Dubai International Convention

& Exhibition Centre, UAE

ITA - AITES WORLDTUNNEL CONGRESS

POSTER PAPERPROCEEDINGS

1

Energy Tunnel Segmental Lining: an Experimental Site in Turin MetroMarco Barla1 and Alessandra Insana21 Dept. of Structural, Building and Geotechnical Engineering, Politecnico di Torino, Corso Duca degli Abruzzi, 24, 10129 Turin, Italy, marco.barla@polito.it 2 Dept. of Structural, Building and Geotechnical Engineering, Politecnico di Torino, Corso Duca degli Abruzzi, 24, 10129 Turin, Italy, alessandra.insana@polito.it

ABSTRACTEnergy geo-structures such as piles, diaphragm walls, tunnel linings and anchors are rapidly spreading in Europe and all around the world. They differ from traditional geotechnical infrastructures because they are equipped with absorber pipes. The working fluid circulating in the pipes extracts or injects heat into the ground depending on the season in order to heat or cool, respectively, the surrounding buildings or de-ice infrastructures. Most current practical applications are related to energy piles and retaining walls, but some examples of energy tunnels can be found in recent literature. Tunnels are particularly suited for this kind of application thanks to the large volume of ground involved and the wide heat exchange surface. Today, a large number of urban tunnels are constructed using tunnel boring machine (TBM) and segmental linings. This is also the case for Turin Metro Line 1, which is currently being extended southward, from Lingotto to Piazza Bengasi (1.9 km of tunnel length and 2 stations). Previous studies demonstrate the presence of favorable underground water flow conditions for the application of low enthalpy geothermal technology.

An experimental site consisting of two thermally activated lining rings was designed by the authors according to the novel ENERTUN patent to be installed during construction of this new section of Turin metro. Pipes were placed in each segment in a meandering fashion, along the circumferential direction, perpendicularly to the tunnel axis. A number of sensors were integrated in the segmental lining before pre-casting, with the aim of measuring strains, stresses and temperatures. Moreover, temperature probes were located inside the tunnel. After monitoring the background in-situ conditions, activating and verifying the correct operation of the installation, heating mode test and cooling mode test are being performed to investigate the efficiency of the thermal lining and validate predictive numerical analyses against real in-situ data.

This paper is intended to describe the characteristics of the experimental site and present some early results collected compared with thermo-hydro finite element analyses.

Key Words: Geothermal Energy, Energy Tunnel, Experimental Site, Monitoring.

1. INTRODUCTION

The global need for clean and renewable energy sources is growing, since they represent one of the most effective tools against climate change. In this context, shallow geothermal energy can represent a valid way to go, given that it is accessible everywhere in the world (below 10-15 m up to 400 m depth). Today, the use of shallow geothermal energy is common to many countries all around the world. Usually, heat is extracted from the ground for domestic heating or to produce hot water by means of closed or open loop systems.

2

Since the 1980s, a number of underground geotechnical structures (piles, diaphragm walls, tunnel linings) have been turned into heat exchangers, the so-called energy geostructures, by embedding absorber pipes in their concrete mass (Laloui and Di Donna, 2013; Barla and Di Donna, 2016). A carrier fluid circulating along the pipes extracts or injects heat from or into the ground with the aim of heating/cooling buildings respectively. The integration of a structural element together with an energetic purpose requires little effort, though resulting in great environmental and economic benefits. Most practical applications are related to energy piles and diaphragm walls and are already operational especially in Austria, UK, Germany and Switzerland (Brandl, 2006). Feasibility studies and an increasing number of new applications are under way.

Recently, an interest in applying this technology to tunnels has emerged, as witnessed by the number of studies that investigate thermal exploitation through tunnels (Barla and Perino, 2014; Wilhelm and Rybach, 2003; Adam and Markiewicz, 2009; Schneider and Moormann, 2010; Frodl et al., 2010; Franzius and Pralle, 2011; Lee et al., 2012; Zhang et al., 2013; Nicholson et al., 2014; Barla et al., 2016; Di Donna and Barla, 2016; Moormann et al., 2016). Compared to building foundations, an advantage of turning tunnels into energy tunnels is the larger volume of ground and surface for heat exchange. Furthermore, when mechanized tunnelling is used, the thermal activation process can be optimized in factory.

After a brief introduction on energy tunnel design aspects, this paper will focus on the description of the experimental site set up and its implementation in the construction site. Some preliminary results will be presented and discussed.

2. ENERGY TUNNELS

In general, a geotechnical structure can be turned into an energy geostructure by placing PE-Xa pipes into its concrete body. The fluid flowing in the pipes represents the carrier to transfer heat from the ground to the buildings or vice versa thanks to heat pumps. Heat pumps utilize a working fluid called refrigerant, that is circulated using a compressor, which drives the process.

Thermoactive geostructures couple a structural role with an energetic one, subsequently the operational depth is driven by the location and size required for the structural design. What is essential for the thermal activation purpose, is a nearly constant ground temperature all year round. Despite the seasonal fluctuations concerning the first 5-8 m, in most continental Europe the ground temperature varies between 8°C and 16°C, but remains constant the whole year, up to about 50 m. This is the depth range where most geostructures can be found. Therefore, the ground can operate both as a heat source/sink during winter/summer respectively: in the first case, the ground temperature will be warmer than the external air temperature, whilst it will be cooler in the second case.

Depending on the tunnel excavation method, different technologies are already available to instrument the tunnel with a circuit of pipes. If conventional tunnelling method (e. g. New Austrian Tunnelling Method, NATM) is adopted, absorber pipes can be attached to non-woven geosynthetics off site and then placed between the primary and the secondary lining, thus making prefabrication possible and providing for easier in situ installation. The first application of this kind can be found in the Lainzer tunnel in Austria (Adam and Markiewicz, 2009). However, today most urban

3

tunnels are constructed by using tunnel boring machines (TBM) and prefabricated segmental linings. Therefore, in the case of mechanised excavation, the segments can be prepared in factory by embedding nets of pipes for heat exchange before casting (Frodl et al., 2010; Franzius and Pralle, 2011; Nicholson et al., 2014; Barla et al., 2016). With regard to standard segments, special coupling pockets need to be envisaged and specifically designed. Pockets have the fundamental role of allowing pipes exiting each segment to be connected and to form a continuous circuit. If structural design occurs in parallel with the hydraulic setup planning, it should not be difficult to include pockets. Each ring is usually made of 6-7 segments. To reduce head losses, not all rings have to be connected to the main conduit, but groups of them can be connected in parallel forming smaller circuits that, in turn, are connected to the main conduit. An advantage of connecting rings in parallel is the possibility of excluding some of them in case of unexpected glitches. Examples of precast energy segments can be found in Austria, where 27 rings (corresponding to a length of 54 m) in Jenbach Tunnel were thermally activated to satisfy Jenbach town council energy demand (Frodl et al., 2010; Moormann et al., 2016). The system was previously tested in a field trial in 2007, by activating few rings of a high-speed railway tunnel in Germany (Franzius and Pralle, 2011). A comprehensive monitoring system composed by over 40 gauges to read temperatures within the segments, the tunnel air, segment surfaces and up to 7.5 m into the surrounding ground was used. Activation took place in 2009, prior to tunnel opening, using a temporary heat pump. Despite these attempts, none of them managed to spread the application of this technology in order to make it a proven, sound and smart utilization of tunnels infrastructures against the climate change challenge.

A novel energy tunnel precast segmental lining (ENERTUN) was designed and patented (Barla and Di Donna, 2016; Barla and Di Donna, 2017) at Politecnico di Torino. Compared to previous configurations, ENERTUN suggests an innovative and more efficient layout of the net of pipes. In fact, they are arranged so as that their main direction results to be perpendicular to the tunnel axis, thus reducing bends and, subsequently, head losses by about 20-30% on one hand. On the other hand, holding true the same material properties, geometry and ground conditions, thermal exchange efficiency increases by 5-10%, particularly in those situations where groundwater flow is perpendicular to the tunnel axis. Moreover, according to the purpose, the patent proposes three different configurations. The ground configuration (Figure 1a) is conceived to exchange heat with the ground, the air configuration (Figure 1b) allows to operate on the air inside the tunnel, while, lastly, the ground&air configuration (Figure 1c) is a combined solution that fulfils both previously mentioned tasks.

(a) (b) (c)

Figure 1. Possible configurations of the ENERTUN energy tunnel precast segmental lining: (a) ground, (b) air and (c) ground&air.

4

Thermo-hydro numerical analyses carried out to investigate the energy efficiency of thermal activation of tunnels and their influence on the surrounding ground demonstrate the advantageous conditions existing in Turin, related to the favorable underground water flow, with regards to other case studies (Franzius and Pralle, 2011). In fact, the exchanged heat was computed to be 53 and 74 W per square meter of tunnel lining during winter and summer respectively and the difference in temperature between inlet to and outlet from the primary circuit (ground side) was between 3 to 4 °C in both cases (Barla et al., 2016). Numerical results were so promising to encourage the Authors to implement a real scale installation in Turin, by taking advantage of the Metro Line 1 South Extension now under construction.

3. DESCRIPTION OF THE TURIN METRO LINE 1 SOUTH EXTENSION

Turin Metro Line 1 connects the North-West part of the city with the South-East one, from Fermi station to Lingotto multifunctional centre, for a total length of 13.4 km and 21 stations. The construction started in 2000, in view of Turin 2006 Winter Olympic Games. In 2006 the first section from Fermi to XVIII Dicembre was opened, immediately followed by the second one up to Porta Nuova in 2007. The last part of the line towards Lingotto was completed in 2011. Two additional stations located at the Southern border of the city, Italia ’61, planned to serve the future Piedmont Region Headquarters, and Bengasi, are currently under construction (Figure 2).

The Lingotto-Bengasi section is characterized by a 1.9 km length, 2 intermediate ventilation shafts (PB1 and PB2), a terminal shaft of end section (PBT) located approximately 200 m beyond Bengasi station and an intersection to allow for the future construction of the branch line to Lingotto railway station. The terminal tunnel will allow the inversion of trains behind Bengasi station (as it happens today at Fermi), in the section between the station itself and the terminal shaft. This solution will make it possible to fully exploit both the station (currently only one platform is used at Lingotto) and the potential of the VAL system (automatic light vehicle): at peak times it will be possible to guarantee a train frequency every 69 seconds. Additionally, the length of the concerned section enables parking of four trains, ready to get in service depending on the users flows fluctuations.

At the time of writing the shielded EPB (Earth Pressure Balance) TBM has completed the excavation of the terminal tunnel and of the 834-m long section linking Bengasi to Italia ’61, for a total length of 1043 m. After a short period of maintenance, the TBM has started excavating again towards Lingotto, where it will be dismantled and extracted. It is 100 meters long and has a diameter of 7.70 m. Moving its more than 400 tons of weight, it is able to dig into the groundwater as excavation takes place under the water table whose depth lies in the range from 12 to 13 m. The average depth of the tunnel is about 20 m. The tunnel lining is 30 cm thick and is made of 6 precast concrete segments mounted by the TBM itself while excavating and permanently sealed to protect tunnel tube against groundwater. Turin metro tunnel can be classified as a cold tunnel, that is internal air temperature is similar to ground surface temperature and thermal influence due to fast-moving trains operation is neglectable.

5

Figure 2. Turin Metro Line 1 layout. South Extension is highlighted in the box.

4. DESCRIPTION OF THE EXPERIMENTAL SITE

4.1. Location

The experimental site has been installed in the tunnel of Turin Metro Line 1 South Extension under construction, 40 m northwards from Bengasi station (Figure 3). Installation chainage was decided in accordance with the construction site managers to minimize impact with the construction operations. The prototype of energy tunnel consists of two rings (number 179 and 180, as shown in Figure 3) equipped with the ground&air configuration, for a total longitudinal length of 2.80 m. Each ring is composed by five segments plus a key and all of them were thermally activated. The “control room” (red circle in Figure 3) is represented by an open space located close to the future elevator shaft (at present stairs are located here to allow access to the tunnel entrance). In this space, the ground&air hydraulic circuit starting and ending points can be found and management of the heat pump and of the data acquisition system are possible without interfering with the muck train going back and forth from the TBM backup in the tunnel.

6

Figure 3. Experimental site location: activated rings, hydraulic circuit and heat pump.

4.2. Prototype design

Two rings of segmental lining were fully equipped with the ground&air net of pipes to exchange heat, for a total of 12 energy segments. As pipes configuration design took place after reinforced concrete design, their exact location and spacing was dictated by the rebars position. For example, pipes configuration could not be symmetrical with respect to the segments axis, as here two pick-up pins are located from which segments are grasped by the erector mechanical pick-up unit. The two nets of pipes, one close to the extrados, the other close to the intrados will allow to test alternatively all three different configurations.

The pipes used are fabricated in reticulated polyethylene (PE-Xa), have an external diameter of 20 mm and a thickness of 2 mm, with the exception of flow and return pipes (32 mm external diameter, 2.9 mm thickness) that collect heat carrier fluid coming from both rings working in parallel. They were tied to rebars by hand in the precast concrete plant. Of course, this process was time-consuming if compared to the production of a standard segment, but can be easily optimized in the case of an industrial scale installation.

The ground&air nets of pipes in each segment need to be connected to the adjacent segments’ ones once the TBM has mounted them on site. To do so, four special coupling pockets were included at the intrados. From here, pipes can exit the segment and be handled for the following connection. At the end,

7

two continuous circuits are available, one at the extrados and the other at the intrados. The geometry of pipes is shown in Figure 4 for a sample segment, both schematically and in real views.

Figure 4. Example of pipes geometry and monitoring system configuration for segment K: (a) ring 179 and (b) ring 180.

Air pipes are depicted in red, ground pipes in blue.

4.3. Monitoring system

A comprehensive monitoring system was installed to monitor the energy tunnel performance both from a thermal and a structural point of view.

The two ENERTUN rings were instrumented with a specifically designed monitoring system to monitor stresses, strains and temperatures in the lining. The sensors detailed location is shown in Figure 5.

Figure 5. Comprehensive monitoring system design:

(a) ring 179 and (b) ring 180.

8

On one hand, ring 179 was equipped with pressure cells to monitor radial (segments B, D), hoop (segments A, C, E) and longitudinal (segment K) stresses and with intrados and extrados vibrating wire strain transducers to monitor hoop (segments B, D) and longitudinal (segment K) strains. Both pressure cells and vibrating wire strain transducers, supplied by Sisgeo (Italy), are also able to provide a local temperature measurement thanks to an embedded NTC thermistor. On the other hand, a monitoring system based on single-mode fiber optics sensors was installed in ring 180. These sensors are able to measure strains and temperatures at various locations (segments B, D, K intrados and extrados). Measurements take place thanks to an interrogation system. Wavelengths are measured with time, therefore a post-processing step is needed to convert them into quantities of physical significance, such as temperature and strain. Both rings are provided with multimodal fibers, that would be able to return distributed measurements, all along their length. In Figure 4 an example of the monitoring system for segment K can be seen: on the top left, in green is the pressure cell and in purple the vibrating wire strain transducer, whereas on the bottom left in blue is the single-mode fiber optics system, and in both in orange is the multimodal one. On the right, some details of the sensors and coupling pockets are magnified.

The upstream groundwater temperature is continuously monitored thanks to a temperature probe located inside a piezometer well within the construction site (Figure 6a,b,c). Tunnel air and segments surface temperature are also monitored by using additional probes (Figure 6d).

4.4. Installation

The energy segments were casted at the precast concrete plant in May 2017 (Figure 7a-d). After curing, circulation tests were carried out to ensure pipes integrity (Figure 7e). Then, they were transported to the construction site (Figure 7f) and placed on site by the TBM at the beginning of July 2017. As of this date, sensors electrical connections and hydraulic connections between adjacent segments were completed (Figure 7g,h) by taking advantage of the TBM backup to reach the highest, hardly accessible points. As a result, two circuits per each ring were obtained, air and ground. As previously mentioned, the two rings were connected in parallel after segments installation and to the heat pump for pump flow and return to achieve the complete ground&air circuit (Figure 7i).

The 6-kW heat pump system represents the fluid circulation and thermal power application device. The heat transfer fluid is a propylene glycol mixed with water allowing to work down to a temperature of -20°C. The system makes it possible to simulate summer and winter heating and cooling cycles. Because of the experimental nature of the project, the secondary circuit of the heat pump is represented by a fan coil unit located close to the heat pump. Therefore, at present there is not a real end user benefitting from the tests, but the heat is dissipated in the air.

4.5. Experimental campaign

The aim of the experimental campaign is to evaluate efficiency, reliability and impact on the lining and the surrounding ground due to the prototype thermal activation. Monitoring started in September 2017 with the assessment of undisturbed conditions at the site. After this initial phase, whose purpose was also to verify the regular functioning of the system, winter mode tests were simulated and are

9

currently ongoing (December 2017). The experimental programme includes cooling mode tests too, but these will be performed later to reproduce summer conditions.Heating mode tests started at the end of October 2017 and concerned both the air and the ground circuit separately. Ground cooling occurs in this phase. At the end of each test the heat pump is turned off for long enough to ensure returning to the initial undisturbed thermal and mechanical conditions. Data acquisition occurs remotely in real time. For each test the following quantities are recorded: (i) inlet and outlet ground-side (ENERTUN or primary circuit) temperatures, (ii) inlet and outlet virtual user-side (fan coil unit or secondary circuit) temperatures, (iii) temperatures and stress-strain state in the lining, (iv) temperatures inside the tunnel, (v) upstream groundwater temperature, (vi) external temperature. Monitored data are periodically downloaded, processed and interpreted. What can be controlled by the user is the return temperature from the secondary circuit. The heat carrier fluid inlet temperature to the ENERTUN circuit and its velocity are automatically adjusted by the device. From every test the thermal power exchanged with the ground is computed and the mechanical thermally-induced effects in the lining are investigated.

Figure 6. Additional temperature probes: (a) piezometer location within the construction site, (b) relative temperature node and (c) borehole, (d) tunnel

air and segments surface node.

10

Figure 7. Installation stages: (a) Reinforcement cage and embedded circuits, (b) mould preparation, (c) casting, (d) concrete curing,

(e) circulation tests, (f) energy segments transported to the construction site, (g) hydraulic circuit before and after the connection, (h) Western

springline view of the completed ENERTUN rings, (i) final circuit

As a preliminary example, Figure 8 displays the outputs (i), (ii) and (vi) of a heating mode test, carried out for the ground circuit (both rings 179 and 180) at a target temperature of 30 °C (test code 171106_G_H_T30_179180). The test lasted for slightly more than two days and its interpretation allowed to draw some conclusions about thermal energy that can be extracted from the ground. The heat Q (expressed in Watts) extracted during the test can be computed using Eq. (1) below:

Q = mcw |To - Ti | (1)

where m is the mass flow rate (0.22 kg/s in this test), cw is the water heat capacity (4200 J/(kg·K)), Ti and To are the recorded inlet and outlet temperatures of the

11

pipe circuit, respectively. Following Eq. (1), the total extracted power is 1.59 kW, equivalent to 48.70 W/m2 and 1132.25 W/m. The area under the total extracted power-time curve represents thermal energy extracted, equal to 97.70 kWht. Based on the electrical energy used to operate the heat pump (35.83 kWhe), the resulting COP (Coefficient of Peformance) is 2.73. As regards the inlet and outlet temperature difference with respect to the ENERTUN ground circuit, the maximum obtained value is 4.5 °C, higher than the corresponding simulated result from coupled thermo-hydro finite element analyses, between 3 to 4 °C shown by previous studies (Barla et al., 2016; Di Donna and Barla, 2016). Hence, the numerical model reliability is confirmed, although the model is not exactly coherent with the configuration of the experimental installation, for example in terms of geometry and working conditions, and the results corroborate the efficiency of the technology.

Figure 8. Temperature outputs of a ground heating mode test (test code 171106_G_H_T30_179180)

5. CONCLUSIONS

A very challenging experimental site has been implemented in the tunnel of the Turin Metro Line 1, South Extension. Two complete rings of thermally active segmental lining were installed, by using the technology of the ENERTUN segment (patent priority number: 102016000020821). The aim is to evaluate the efficiency of the system, both from the energy, structural and geotechnical point of view.

The testing programme is currently under way: a number of winter mode tests have already been carried out, but they will continue during the whole winter season. Afterwards, the system will be switched to the summer operational mode. Monitoring tools, such as pressure cells, vibrating wire strain transducers, thermistors, fiber optics strain and temperature sensors embedded in the concrete lining are used to evaluate the thermo-mechanical behavior of the system during testing. Preliminary experimental results are very promising and in line with existing thermo-hydro numerical models. Further study is needed to verify this trend under different operating conditions of the system.

Future work is planned to build a new thermo-hydro numerical model, able to reproduce accurately geometric, thermal and hydraulic conditions at the site. Moreover, a 3D thermo-mechanical model will allow to back-analyse experimental

12

results and to extend them to different geometries and environments, also in view of future applications such as that of Turin Metro Line 2, now at the stage of preliminary design.

6. ACKNOWLEDGEMENTS

The Authors wish to acknowledge the contribution of Politecnico di Torino and Compagnia di San Paolo ‘Proof of concept’ funding that allowed covering the expenses of the experimental site and the fruitful cooperation of InfraTo and Consorzio Integra, responsible for the construction of the Metro Line 1 South Extension. Moreover, Authors are grateful to the support received by CMC (Cooperativa Muratori Cementisti), AK Ingegneria Geotecnica srl, Sisgeo, AR.CO green energy, Winet Srl, Italia 61 scarl, also involved in the project. Finally, Authors are also very thankful for the help received by Alice Di Donna, Giorgio Giannì and Fabrizio Zacco in different stages of the project.

7. REFERENCES

Adam, D., and Markiewicz, R. (2009). Energy from earth-coupled structures, foundations, tunnels and sewers. Géotechnique, 59(3):229–36.

Barla, M., and Di Donna, A. (2016). Editorial Themed issue on energy geostructures. Environmental Geotechnics, 3(4):188–9.

Barla, M., and Di Donna, A. (2016). Conci energetici per il rivestimento delle gallerie. Strade e autostrade: gallerie&tunneling, (4):2–5.

Barla, M., and Di Donna, A. (2017). Energy tunnels for heating and cooling, EURO:TUN 2017, 18-20 April 2017, Innsbruck University, Austria.

Barla, M., and Perino, A. (2014). Energy from geo-structures: a topic of growing interest. Environmental geotechnics, 2(1):3–7.

Barla, M., Di Donna, A., and Perino, A. (2016). Application of energy tunnels to an urban environment. Geothermics. Geothermics, 61:104–113.

Brandl, H. (2006). Energy foundations and other thermo-active ground structures. Géotechnique, 56(2):81–122.

Di Donna, A., and Barla, M. (2016). The role of ground conditions and properties on the efficiency of energy tunnels. Environmental geotechnics, 3(4).

Franzius, J.N., and Pralle, N. (2011). Turning segmental tunnels into sources of renewable energy. Proceedings of the ICE - Civil Engineering, 164(1):35–40.

Frodl, S., Franzius, J.N., and Bartl, T. (2010). Design and construction of the tunnel geothermal system in Jenbach./Planung und Bau der Tunnel-Geothermieanlage in Jenbach. Geomechanics and Tunnelling, 3(5):658–68.

Laloui, L., and Di Donna, A. (2013). Energy geostructures: innovation in underground engineering. ISTE Ltd and John Wiley & Sons Inc., 320 p.Lee, C., Park, S., Won, J., Jeoung, J., Sohn, B., and Choi,

13

H. (2012). Evaluation of thermal performance of energy textile installed in Tunnel. Renewable Energy. Elsevier Ltd, 42:11–22.

Moormann, C., Buhmann, P., Friedemann, W., Homuth, S., and Pralle, N. (2016). Tunnel geothermics - International experience with renewable energy concepts in tunnelling / Tunnelgeothermie - Internationale Erfahrungen zu regenerativen Energiekonzepten im Tunnelbau. Geomechanics and Tunnelling, 9(5):467–80.

Nicholson, D.P., Chen, Q., de Silva, M., Winter, A., and Winterling, R. (2014). The design of thermal tunnel energy segments for Crossrail, UK. Engineering Sustainability, 167(3):118–34.

Schneider, M., and Moormann, C. (2010). GeoTU6 – a geothermal Research Project for Tunnels, 14–21.

Wilhelm, J., and Rybach, L. (2003). The geothermal potential of Swiss Alpine tunnels. Geothermics., 32(4):557–68.

Zhang, G., Xia, C., Sun, M., Zou, Y., and Xiao, S. (2013). A new model and analytical solution for the heat conduction of tunnel lining ground heat exchangers. Cold Regions Science and Technology. Elsevier B.V.; 88:59–66.

FOR VISITINGTHANK YOU

ITAAITES