PSI - Issue 78

Eugênio Moreira et al. / Procedia Structural Integrity 78 (2026) 1967–1974

1968

1. Introduction

Ancient medieval walls in historic cities hold significant architectural, historical, and urban value, serving as en during symbols of local identity and collective memory (Bruce and Creighton, 2006; Tu¨lek and Atik, 2014). Their preservation, as emphasized by UNESCO World Heritage criteria, is crucial to sustaining a sense of place for commu nities. Structurally, these walls typically comprise external stone or brick leaves enclosing a core of loose rubble and weak mortar, which creates inherent vulnerabilities: inferior mechanical properties of the internal core, deteriorated mortar joints, and insu ffi cient connections between wall leaves (Mammoliti et al., 2021). During seismic events, such construction leads to poor performance under horizontal forces, with the separation of external and internal layers and pronounced out-of-plane deformations representing common and critical failure modes (Candela et al., 2016; Corradi et al., 2017). Some multi-leaf walls, particularly those with irregular small stones and degraded mortar, are prone to complete disaggregation, resulting in collapse through loss of confinement and rapid debris fall, an e ff ect exacerbated by seismic events with substantial vertical acceleration components (Borri et al., 2020). Current seismic assessments often overlook the influence of daily thermal cycles, which continually modify stress states, potentially predisposing walls to specific damage patterns during seismic activity. Addressing this gap, the present study introduces an integrated thermal-mechanical modeling approach applied to Gubbio’s historic wall, ex plicitly examining thermal pre-conditioning as a factor in seismic vulnerability. The methodology combines baseline Structural Health Monitoring (SHM) with Finite Element Modeling (FEM) using the Concrete Damage Plasticity (CDP) framework, integrating thermal and mechanical loading histories. Five simulated scenarios are considered: baseline thermal-mechanical behavior without seismic input; the response to a real seismic event following thermal pre-conditioning; and three further cases with amplified seismic magnitudes to capture damage progression and altered structural mechanisms. The objectives are to investigate baseline thermal mechanical conditions using realistic environmental data; to develop a coupled numerical model with CDP; to evaluate seismic vulnerability through nonlinear dynamic analysis of thermally pre-conditioned masonry; and to quantify post earthquake performance. By aligning continuous monitoring data with advanced numerical simulation, this research advances understand ing of thermal-seismic interactions in heritage masonry. The proposed framework contributes not only to improved seismic risk assessment but also to digital twin applications supporting real-time damage detection and conservation planning, reflecting the complex interplay between environmental and seismic loading in architectural heritage. The Concrete Damage Plasticity (CDP) model is widely adopted to simulate inelastic behavior and failure mech anisms of multi-leaf stone masonry under seismic and complex loadings, capturing both tensile cracking and com pressive crushing (Valente and Milani, 2019). CDP’s combined damage-plasticity formulation enables realistic failure prediction for masonry, outperforming earlier linear or elastic damage models, especially for repeated or cyclic loading (Rainone et al., 2023). Despite its strengths, CDP application to heritage masonry presents challenges, chiefly in parameter calibration for masonry-specific material properties and interface behaviors (Rainone et al., 2023). Accurate modeling of multi leaf masonry requires consideration of discrete debonding, slip, and separation at interleaf interfaces, which CDP’s isotropic damage laws may not fully capture, potentially underestimating interfacial failure modes (Amer et al., 2021). Most published CDP studies have addressed mechanical loading. At the same time, integration of thermal pre conditioning remains limited, even though thermal cycles can significantly a ff ect stress states and material properties before seismic events. Mesh sensitivity, convergence issues in nonlinear regimes, and computational demands further complicate CDP use for detailed heritage analyses (Rainone et al., 2023). Recent developments have focused on parameter calibration and validation, with advanced FEM analyses demon strating CDP’s ability to predict masonry behavior within 3–9% peak load error when well-calibrated (Dewi et al., 2025). Experimental validation, including shake-table and modal analysis, as well as sensitivity studies on critical parameters (e.g., dilation angle, viscosity, mesh size), have provided practical guidance (Miglietta et al., 2021; Dewi et al., 2025). 2. Background and Literature Review