National Technical University of Athens (NTUA)
Algorithmic thermophysical analysis for composite PCM-perlite materials, using machine learning mechanisms, for optimizing passive energy storage in buildings
Abstract
dc:descriptionThe European Commission has set climate neutral targets to reduce carbon emissions by 40% by 2030, while in parallel increase the penetration of Renewable Energy Sources (RES) up to 32% and enhancing energy performance by 32.5%. This has been quantified under the European climate law that expects the EU countries to overall minimise their carbon-related emissions by 55% by 2030. Investigating the benefits deriving from shifting electricity demand in buildings highlights the potential for exploiting the available thermal energy storage capacity resulting from the envelope mass of a building by applying pre-heating or pre-cooling strategies. This concept introduces a passive energy storage system named “structural thermal energy storage” which employs the mass of the structural elements—i.e., walls, slabs and ceilings—to store thermal energy and retrieve it at a later time. The efficiency of this strategy has been proven for ages, with a typical case being the passive night cooling of heavyweight buildings as a traditional way to treat the heated thermal mass of stone or concrete in hot climates. Several studies have investigated the opportunity to increase building envelope thermal mass by utilising composite materials, through incorporating low-cost carriers, like industrial minerals, for absorbing phase change materials (PCMs). PCMs provide a way for exploiting the latent heat when storing thermal energy. In principle, any material that undergoes a phase change process when absorbing or releasing thermal energy can be used as Latent Thermal Energy Storage (LTES). A fundamental characteristic of LTES materials is that there is a phase change temperature at which the material changes from one state (solid or liquid or gas) to another without a change in chemical composition. However, the integration of such composites in building envelope like cement-based mortars, or loose fill applications should maintain desired properties like workability, durability, hydration reaction time, ensuring also competitiveness in relation to environmental footprint and cost. In this framework, recent studies have been dedicated on the utilisation of lightweight industrial minerals as potential carriers due to large surface area, porous structure, low cost and chemical inertness. Τhe effective activation of structural thermal energy storage corresponds with the physical and thermal properties tailored to each building application. On this account, there is a need for analysing the thermal behaviour of hybrid LTES-enhanced composites and evaluate the heat transfer phenomena occurring between the building’s envelope layers for each use case. Furthermore, for each layer, like mortars and plasters, the thermal interaction between the particles of binder and fillers, within a volume unit and the packing density is crucial. Particle morphology of binders’ effects on binder-filler interfacial interaction, while packing characteristics of filler particles could provide insight to possible link between particle morphology and interfacial behaviour. Hybrid LTES-enhanced composites is a promising additive that can replace conventional fillers for storing energy in thermal mass, as soon as desired end product properties are ensured. LTES-enhanced composites performance is correlated with the carrier ability to capture a melted LTES material within its pores and is affected mainly by the density, the particle size specific surface and porosity. Mixing conditions like temperature (affecting LTES viscosity), duration and pressure are also affecting the sorption process's efficiency and dynamics. All these factors can be modelled for the analysis of LTES sorption into porous industrial minerals and to predict the thermal storage capacity of LTES-enhanced composites but also to evaluate possible utilisation schemes at building envelope. Even more, the combination of mathematical modelling (knowledge-driven) and machine / deep learning (data-driven) are capable of handling complex, high-dimensional data with numerous parameters. The use of physics-informed neural networks, where mechanistic equations guide the learning process allows for more precise calculations, able to successfully predict the thermal behaviour of composite LTES-enhanced materials and their thermal efficiency at building envelope level. This doctoral thesis develops an integrated model for evaluating energy storage capabilities and energy flow of building envelope layers incorporating LTES-enhanced composites in mortars and loose fill applications, at two levels:1.Theoretical and numerical framework:•Thermodynamic prediction of the thermal energy able to be stored in each LTES-enhanced building envelope layer •Kinetics of thermal charging/discharging cycles of the energy stored in LTES-enhanced building thermal mass •Overall efficiency of the technology, promoted for two envelope building applications: loose fill and mortar utilising as filler industrial mineral carrying PCM enhanced composites •LTES-enhanced mortar’ thermal mass efficiency, by analysing thermophysical filler properties, and by ensuring in parallel end product compatibility for building applications •Thermal interaction between the particles of binder and fillers of mortar, within a volume unit in relation to the packing density •Packing density correlation with particle size distribution •Sorption capacity of porous structure, in relation to density, particle size distribution, specific surface and porosity. Sorption capacity is quantified in both inter capacity (open cavities of the porous structure) and intra capacity in relation to the open pores. •Sorption mixing process efficacy in relation to the mixing temperature affecting also contact angle, interfacial surface tension, LTES material density and dynamic viscosity 2.Data driven analytical modules: •SEM image recognition and analysing particles morphology such as the circular shape factor, the porosity, the specific surface, the radius of surface and their size distribution, including the mean radius and the standard deviation. The correlation of inter capacity and the image analysis derived properties the calculation of the volume available to sorb melted PCM in particles surface •Particles 3D visualisation based on SEM image recognition module taking into account particle size distribution, for calculating packing density •Computational tool for converting volume weighted particle size distributions to probability distribution. This tool predicts the number of the particles in relation to their size than can be distributed within an inter sorption volume unit This twofold analysis delivers a holistic hybrid data-driven mechanistic model enabling multiscale material and process insight by defining and predicting thermal mass utilisation potential at both particle and building envelope level, evaluating LTES-enhanced composites thermodynamic efficiency to be applied for hybrid energy storage. Accordingly, materials were evaluated based on their energy storage and release capabilities. An organic LTES material, like Rubitherm RT27, with latent heat capacity of 150-200 kJ/kg, was selected for its high energy storage properties combined with compatible to the carrier properties and market-ready accessibility. The LTES-enhanced composites’ specific heat capacity, found to be in the range of 35-40 kJ/kg·K, also facilitating the enhancing of TES applications during heating and cooling cycles. Simulations showed that incorporating LTES-enhanced mortars into building walls could reduce energy consumption by prolonging the thermal discharge of the building block up to 18 and 20 hours. This demonstrates a significant delay of roughly 15 hours compared to a conventional building block being thermally discharges after around 3 to 5 hours. The delayed solidification process effectively prolongs the wall's ability to maintain a stable temperature, as the PCM absorbs heat from the surrounding layers before fully solidifying. This contributes to the extended cooling period observed in the enhanced wall, reducing the risk of sudden temperature drops and enhancing indoor thermal comfort. This reduction was mainly due to the materials' ability to stabilize indoor temperatures, thereby decreasing the reliance on HVAC systems. Furthermore, sorption capacity models analysed factors such as particle size, surface area, and pore volume, particularly highlighting the performance of EP as a PCM carrier due to its high porosity (around 90%) and large surface area (up to 20 m²/g), which allowed for improved energy storage density and heat transfer. The hybrid data-driven mechanistic model developed provides a valuable tool for evaluating the emerging class of next-generation LTES solutions. The multidimensional efficacy developed is overcoming the persistent challenges in material synthesis, scalability, and long- term stability. By encouraging solutions for tool-based evaluation of next-generation LTES materials and through synergies with other matching technologies and promote new LTES technological solutions able to be applied in wider fields encompass spacecraft, photonics, paint emulsions, biomedical and healthcare, textiles, foods, cold chain logistics, as well as energy storage and conversion.
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2025
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Politi, Chrysa
- Πολίτη, Χρύσα
Subjects
dc:subject × 18- Ρόφηση
- Σύνθετα υλικά
- Υβριδική αποθήκευση ενέργειας
- Μηχανιστικά μοντέλα πρόβλεψης
- Sorption
- Composite materials
- Hybrid energy storage system
- Mechanistic predictive models
- Άλλες Επιστήμες Μηχανικού και Τεχνολογίες
- Επιστήμες Μηχανικού και Τεχνολογία
- Μηχανική, διεπιστημονική προσέγγιση
- Μηχανική Υλικών
- Επιστήμη υλικών, διεπιστημονική προσέγγιση
- Other Engineering and Technologies
- Engineering and Technology
- Engineering, interdisciplinary
- Materials Engineering
- Materials science, interdisciplinary
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/59218