THE ENERGY REFURBISHMENT OF EXISTING BUILDINGS FOR THE GOAL OF THE NET OR NEARLY ZERO ENERGY TARGET TO FRONT THE CLIMATE CHANGE

The research presented in this doctoral thesis, titled "The Energy Refurbishment of Existing Buildings for the Goal of the Net or Nearly Zero Energy Target to Front the Climate Change," addresses the urgent global challenge of climate change through the lens of the building sector’s transformation. As the building stock is currently responsible for approximately 37% of global energy-related carbon dioxide emissions and a significant portion of resource consumption, the study explores the transition from traditional energy paradigms toward Nearly Zero Energy Buildings (nZEB), Zero-Emission Buildings (ZEB), and Positive Energy Buildings (PEB). This research is deeply rooted in the evolving international and European regulatory framework, specifically examining the implications of the European Green Deal, the REPowerEU plan, and the Energy Performance of Buildings Directive (EPBD IV). The central objective of the work is to provide a multi-scale, holistic evaluation of energy refurbishment strategies, focusing on their real-world performance, economic feasibility, indoor environmental quality, and long-term resilience against future climatic shifts predicted for 2050 and 2080. The methodological approach is characterized by a sophisticated integration of diverse scientific tools, ranging from building energy simulation (BES) and computational fluid dynamics (CFD) to advanced artificial intelligence models such as Recurrent Neural Networks (RNN), including Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) architectures. The investigation begins by contextualizing the current environmental crisis, highlighting that anthropogenic activities have led to an unprecedented increase in greenhouse gas concentrations and a consequent rise in global temperatures. Within this framework, the thesis analyzes the Italian legislative landscape, assessing the effectiveness of incentives like the "Superbonus 110%" in accelerating the renovation of existing structures. A significant portion of the research is dedicated to empirical and numerical performance investigations of nZEB designs. Through the analysis of a real case study in Benevento, a Mediterranean climate zone, the study reveals critical gaps in current design practices. Using high-resolution CFD simulations, it demonstrates that even in buildings meeting the highest energy standards, local thermal discomfort can persist due to radiant temperature asymmetries and inadequate control systems. Specifically, the findings suggest that a single thermostat approach in highly insulated envelopes is often insufficient to ensure uniform comfort, particularly during unseasonably warm winter days when solar gains can render active heating unnecessary but lead to localized overheating if not managed by occupant behavior or smart automation. The transition from nZEB to Positive Energy Buildings (PEB) is further explored through the lens of technological integration, particularly regarding the growing demand for electric vehicle (EV) charging. The research highlights a technical paradox where an edifice designed to produce an annual energy surplus may still experience monthly energy deficits during winter months when the load from EV charging is factored in. This necessitates a radical shift in how energy balances are calculated, moving toward a more granular, time-dependent analysis that accounts for the stochastic nature of both renewable generation and user- driven energy demand. Furthermore, the study delves into the socio-technical aspects of energy refurbishment through the REHOUSE project, focusing on the regeneration of social housing in Margherita di Savoia. By proposing a multifunctional "energy exoskeleton" equipped with colored Building-Integrated Photovoltaics (BIPV) and a smart thermal energy hub with phase-change materials (PCM), the research illustrates how aesthetic, structural, and energy goals can be harmonized. The use of the ENVI-met software allows for a detailed assessment of the microclimatic impact of such interventions, proving that the inclusion of green spaces and cool materials is vital to mitigate the Urban Heat Island (UHI) effect, which can otherwise compromise the cooling performance of even the most efficient buildings. A core innovation of this thesis lies in its application of machine learning for climate and energy forecasting. Recognizing that traditional climate files (TMY) often fail to capture recent extreme weather trends, we developed and trained neural network models to predict solar radiation and temperature variations. The comparative analysis of LSTM and GRU architectures shows a high level of accuracy in capturing long-term trends, such as the observed 7.7°C increase in maximum temperatures in the Benevento area over a five- year period. These models provide a robust basis for predicting future energy needs, suggesting a dramatic increase in cooling demand (up to 59% by 2027) that current static design methods fail to anticipate. This leads to the critical theme of resilience. The research tests the performance of current nZEB solutions against the IPCC’s future climate scenarios (RCP 4.5 and 8.5), uncovering what is described as the "insulation paradox" or the "resilience trap." It is revealed that building envelopes optimized for current conditions, particularly those with high levels of thermal insulation and specific glazing types, may become the least resilient in future hotter climates. For instance, while high-performance glazing currently provides significant savings, it shows the highest performance degradation by 2080, leading to a nearly 80% increase in cooling needs. Perhaps the most provocative finding of the research concerns the discrepancy between normative semi- stationary calculation methods (such as those prescribed by Italian UNI/TS 11300 standards) and dynamic hourly simulations (EnergyPlus). The study proves that the semi-stationary approach contains physical inaccuracies regarding the behavior of insulated envelopes in summer. While the normative model suggests that increasing insulation leads to a massive increase in cooling demand by "trapping" heat inside—a conclusion that might discourage energy efficiency measures—the dynamic simulation, which correctly accounts for thermal inertia and surface radiation, shows the opposite. In reality, insulation remains beneficial in summer, reducing the cooling load by 28% to 36% even under future extreme scenarios. This highlights a dangerous risk where policy-making and engineering decisions based on simplified normative tools could lead to sub-optimal or even detrimental design choices. The thesis also extends its analysis to tropical contexts, specifically Santo Domingo, to evaluate the global applicability of the nZEB model. Here, the research emphasizes the impact of physical degradation of photovoltaic systems and the role of high humidity on energy performance. The results suggest that in such climates, the nZEB target is highly volatile and requires a significant "safety margin" or over-dimensioning of renewable sources to maintain compliance over the building's lifecycle. The conclusions underscore that to effectively "front" climate change, buildings must be designed not as static entities but as dynamic systems capable of adapting to a rapidly evolving environment. Ultimately, the thesis demonstrates that while the technology to achieve net-zero goals exists, its successful implementation depends on our ability to predict future stresses and to design for a climate that is significantly different from the one we experience today. This comprehensive study provides a roadmap for researchers, policymakers, and practitioners to navigate the complexities of the energy transition, ensuring that the buildings of tomorrow are not only efficient but also comfortable, economically viable, and truly resilient to the challenges of a warming planet. The findings serve as a call to action to move beyond the minimum requirements of current laws and to embrace a more ambitious, data-driven, and foresight-oriented approach to architectural and environmental engineering. Through the detailed evidence provided across multiple case studies and technological evaluations, this work establishes a rigorous scientific basis for the next generation of energy refurbishment strategies, emphasizing that the path to decarbonization must be paved with both innovation and physical accuracy. In conclusion, the research reaffirms that the energy refurbishment of the existing building stock is not merely a technical necessity but a critical socio-economic imperative that requires a synthesis of information technology, environmental science, and advanced engineering to safeguard the habitability of our urban environments for decades to come. This transition from "nearly zero" to "positive" and "resilient" is the definitive frontier of modern building science, as highlighted by the extensive simulations and analyses conducted throughout this doctoral research.