Analysis of the Deformation Characteristics of Loess Based on Thermal–Mechanical Coupling under an Energy Internet

In response to the major challenges faced by China’s transition to green low-carbon energy under the dual-carbon goal, the use of energy Internet cross-boundary thinking will help to develop research on the integration of renewable clean energy and buildings. Energy piles are a new building-energy-saving technology that uses geothermal energy in the shallow soil of the Earth’s surface as a source of cold (heat) to achieve heating in winter and cooling in summer. It is a complex thermomechanical working process that changes the temperature of the rock and soil around the pile, and the temperature change significantly influences the mechanical properties of natural loess. Although the soil temperature can be easily and quickly obtained by using sensors connected to the Internet of Things, the mechanical properties of natural loess will change greatly under the influence of temperature. To explore the influence of temperature on the stress–strain relationship of structural loess, the undrained triaxial consolidation tests were carried out under different temperatures (5, 20, 50 and 70∘C) and different confining pressures (50, 100, 200 and 400[Formula: see text]kPa), and a binary-medium model was introduced to simulate the stress–strain relationship. By introducing the damage rate under temperature change conditions, a binary-medium model of structural loess under variable temperature conditions was established, and the calculation method of the model parameters was proposed. Finally, the calculated results were compared with the test results. The calculation results showed that the established model has good applicability.

A New Steam Medium Model for Fast and Accurate Simulation of District Heating Systems

District heating effectively meets the heating needs of multiple buildings while consuming less resources compared to individual heating at each building. In U.S. district heating systems, steam is the most common heat transport medium. Simulation of large steam district heating systems requires a computationally efficient and accurate steam model. However, the commonly adopted IF97 water model is not fast enough for district-scale simulations, and its discontinuous thermodynamic property functions have shown to cause simulation problems and sometimes failure. To address these issues, this work introduces a new steam medium model for heating applications with invertible polynomial approximations for specific enthalpy and entropy, which simplify the calculations. Further, we adopt a novel split-medium approach in components for district energy systems, dividing liquid and vapor phases of water into two separate models. This avoids the common numerical challenges at the phase change boundary. We implemented the model in the equation-based Modelica language and evaluated the accuracy and numerical performance across multiple scales: from fundamental thermodynamic properties to complete heating districts of several sizes. The results show that the new model can calculate specific enthalpy within 2.04% of CVRMSE, but with a 39% reduction in computing time. For complete districts, the new implementation has similar accuracy as the IF97 model for evaluating the energy consumption, but at a speed that is 5.6 – 9.3 times faster. Moreover, for the IF97 model, in our district models the dimension of the nonlinear system of equations of the piping network increases linearly in the problem size, but it stays constant with the new model; this is critically important for large scale system simulations. The new steam medium model is available open-source in the Modelica IBPSA and Buildings Libraries.

Computational Modelling for the Optical Properties of Dye Molecules Adsorbed onto Metallic Nanoparticles

<p><b>The study of light scattering by particles has become fundamental and applied interests in the fields of chemistry, biology, and most importantly in physics. In this context, this thesis focuses on understanding the optical properties of dye layers adsorbed onto metallic nanoparticles (NP), which is essential for interpreting the results of plasmon-dye coupling experiments. To model such a system, Mie theory is often used to solve for the exact solution to Maxwell’s equations for spherical homogeneous and isotropic coated NP. The effects of the NP’s plasmon resonances on the optical properties of the adsorbed dye layer have been predicted using an effective medium model, where the dye-layer is treated as an isotropic layer with an effective dielectric function accounting for the dye resonance. However, this isotropic shell model is inadequate as it cannot account for the dye surface concentration and the anisotropy of the optical response of the dye layer.</b></p> <p>In this thesis, we introduce anisotropic effects within Mie theory and develop microscopic models to define effective dielectric functions which explicitly include the dye-concentration effect in the shell model. Combining anisotropic Mie theory with a concentration-dependent effective shell model allows us to form new theoretical tools to model the optical properties of adsorbed dye layers on metallic NPs of spherical shape. With this new refined effective medium model, we are then able to study shell models for elongated particles beyond the quasi-static approximation. This is implemented using the finite element method (FEM) to numerically solve Maxwell’s equations. The FEM implementation is then used to investigate how the NP’s plasmon resonance can be affected by the dye’s orientation and location on the NP’s surface. We show that the orientation and location of the dye molecules on the NP determine how strongly the plasmon resonance is shifted.</p> <p>The results of this work will improve our ability to accurately model the optical properties of anisotropic molecules adsorbed on metallic NPs. This is important in a number of applications including the development of localised surface plasmon resonance (LSPR) sensing and the design of plasmonic devices.</p>

Computational Modelling for the Optical Properties of Dye Molecules Adsorbed onto Metallic Nanoparticles

<p><b>The study of light scattering by particles has become fundamental and applied interests in the fields of chemistry, biology, and most importantly in physics. In this context, this thesis focuses on understanding the optical properties of dye layers adsorbed onto metallic nanoparticles (NP), which is essential for interpreting the results of plasmon-dye coupling experiments. To model such a system, Mie theory is often used to solve for the exact solution to Maxwell’s equations for spherical homogeneous and isotropic coated NP. The effects of the NP’s plasmon resonances on the optical properties of the adsorbed dye layer have been predicted using an effective medium model, where the dye-layer is treated as an isotropic layer with an effective dielectric function accounting for the dye resonance. However, this isotropic shell model is inadequate as it cannot account for the dye surface concentration and the anisotropy of the optical response of the dye layer.</b></p> <p>In this thesis, we introduce anisotropic effects within Mie theory and develop microscopic models to define effective dielectric functions which explicitly include the dye-concentration effect in the shell model. Combining anisotropic Mie theory with a concentration-dependent effective shell model allows us to form new theoretical tools to model the optical properties of adsorbed dye layers on metallic NPs of spherical shape. With this new refined effective medium model, we are then able to study shell models for elongated particles beyond the quasi-static approximation. This is implemented using the finite element method (FEM) to numerically solve Maxwell’s equations. The FEM implementation is then used to investigate how the NP’s plasmon resonance can be affected by the dye’s orientation and location on the NP’s surface. We show that the orientation and location of the dye molecules on the NP determine how strongly the plasmon resonance is shifted.</p> <p>The results of this work will improve our ability to accurately model the optical properties of anisotropic molecules adsorbed on metallic NPs. This is important in a number of applications including the development of localised surface plasmon resonance (LSPR) sensing and the design of plasmonic devices.</p>

Effective Medium Model Applied to Biopolymer Solutions

Studying dielectric properties of heterogeneous systems is challenged by a problem of uncertainty of the ratio between dielectric permittivity of the system and dielectric permittivities of its components. Such ratios can be obtained in some cases using theoretical effective medium models. However, such models have not yet been developed for all the systems possible. Particularly, there is no effective medium model with filamentary inclusions. Such a theoretical model elaborated based on the fundamental principles of electrodynamics of continuous media is suggested in the present work. Any point of a filamentary inclusion with a length that is significantly greater than the thickness can be regarded as being located in a long cylinder-like fragment of the inclusion with stochastic direction of the cylinder axis relative to the external electric field. With this regard, electric field strength and electric induction values were averaged across the entire volume of a two-phase dielectric material. As a result, a model linking the dielectric permittivity of the two-phase system and the dielectric permittivities of both phases was elaborated. The model appears to be highly relevant for studying solutions of biopolymers, such as nucleic acids, fibrillar proteins and protein aggregates, polysaccharides, by means of electrical impedance spectroscopy, dielectric spectroscopy, and terahertz time-domain spectroscopy. The suggested theoretical model was successfully validated on a DNA solution within the terahertz region.