FINITE ELEMENT MODELLING OF ENERGY PILES USING DIFFERENT CONSTITUTIVE MODELS FOR THE SOIL Mouadh Rafai (mouadh.rafai@studenti.unipg.it) Università degli Studi di Perugia – Dipartimento di Ingegneria Civile ed Ambientale Arianna Lupattelli (arianna.lupattelli@studenti.unipg.it) Università degli Studi di Perugia – Dipartimento di Ingegneria Civile ed Ambientale Diana Salciarini (diana.salciarini@unipg.it) Università degli Studi di Perugia – Dipartimento di Ingegneria Civile ed Ambientale ABSTRACT. In this paper a 2D Finite Element model is presented for the simulation of the coupled Thermo-Hydro-Mechanical (THM) processes induced by the operation of energy piles. The model, implemented in the Plaxis 2D Finite Element (FE) code, was used to simulate the behaviour of a well-known instrumented energy pile during a test performed at Lambeth College in London (Bourne-Webb et al., 2009), comparing the results obtained from different constitutive models used to describe the soil behaviour, namely: i) the simple elasto-plastic Mohr- Coulomb (MC) model; ii) the Modified Cam Clay (MCC) model; iii) the Hardening Strain (HS) model. Comparisons with the experimental results show that all the models, if correctly calibrated, are able to realistically reproduce the behaviour of the pile-soil system from a qualitative point of view, while the best prediction was found using the HS model. 1. INTRODUCTION Energy piles are an emerging technology, with several successful applications throughout the world (see, i.e., Brandl, 2006). They allow the exploiting of renewable and clean source of energy for air conditioning of buildings (Amatya et al., 2012; Murphy et al., 2014; Laloui et al., 2019). In urban areas, heating and cooling of buildings represents one of the main sources of energy consumption and one of the main causes of CO 2 production. The low enthalpy geothermal plants equipped with heat pumps provide one of the most efficient and sustainable solutions. These plants exploit the deep foundations as heat exchangers in direct contact with the soil, which is at constant temperature (10-16°C) during the year. Energy piles are ordinary reinforced concrete piles where a system of pipes is installed to allow the circulation of a heat transfer fluid to extract heat from the subsoil during winter, or release heat to the ground in summer. Many recent studies explored the thermodynamic aspects of the process and the thermal efficiency of the system (Cecinato & Loveridge, 2015; Cecinato & Salciarini, 2022), while others paid attention to the geotechnical and structural consequences associated with the cooling and heating processes of the piles, isolated or in groups (Di Donna et al., 2016; Salciarini et al., 2017; Bourne-Webb et al., 2021). Some experimental evidences obtained from instrumented real-scale energy piles subject to mechanical and thermal loading paths (Laloui et al., 2006; Bourne-Webb et al., 2009) indicate that even modest variations in temperature between the pile and the surrounding soil are able to cause significant changes in the state of stress and deformation in the pile and in the soil, particularly when the pile is embedded in a deep rigid layer. The goal of this work is to investigate the effect of using different constitutive models for the soil to reproduce the complex deformation phenomena caused by temperature variations in the vicinity of energy piles, using a FE-THM model implemented in Plaxis 2D. Experimental data collected by Bourne-Webb et al. (2009) were used to evaluate the forecasting capabilities of the models. 2. THE BENCHMARK CASE STUDY The project for a new building of Lambeth College in London (UK), founded on 143 piles, all equipped with U-shaped pipes that allow them to be used as heat exchangers, provided the benchmark case study for the numerical modelling comparison. The experimental study, led by Bourne-Webb et al. (2009), focused on the evaluation of the impact of thermal cycles on the performance of foundation piles and on the analysis of the THM response of the system. The stratigraphy of the site is shown in Figure 1a. The first layer (from the ground surface up to 1.5 m) consists of backfill, followed by terraced alluvial deposits (from 1.5 to 4.0 m), while the underlying layer consists of London clay. The groundwater table is located at a depth of about 3 m from the ground surface. The test pile XI Incontro Annuale dei Giovani Ingegneri Geotecnici. IAGIG 2022. Reggio Calabria, 12 Luglio 2022 © 2022 Associazione Geotecnica Italiana, Roma, Italia, ISBN 978-88-97517-11-5 83