An integrated multiscale approach for characterization of rock masses subjected to tunnel excavation

The design of tunnels must be conducted based on the knowledge of the territory. The longer the structure, the larger the area to be investigated, and the greater the number of surveys and tests to be performed in order to thoroughly examine all the relevant features. Therefore, optimization of the investigation process is strongly required to obtain complete and reliable data for the design of the infrastructure. The fast development of remote sensing technologies and the affordability of their products have contributed to proving their benefits as supports for investigation, encouraging the spreading of automatic or semi-automatic methods for regional scale surveys. Similarly, considering the scale of the rock outcrop, photogrammetric and laser scanner techniques are well-established techniques for representing geometrical features of rock masses, and the benefits of non-contact surveys in terms of safety and time consumption are acknowledged. Unfortunately, in most cases, data obtained at different scales of investigations are only partially integrated or compared, probably due to the missing exchange of knowledge among experts of different fields（e.g. geologists and geotechnical engineers）. The authors,after experiencing such a lack of connection among the results of different surveys concerning tunnels,propose a multiscale approach for the optimization of the investigation process, starting from the regional scale, to obtain the data that can be useful not only for planning more detailed surveys in a preliminary phase, but also for making previsions on the discontinuity sets that are present in the rock masses subjected to excavations. A methodological process is proposed and illustrated by means of a case study. Preliminary results are discussed to highlight the potentiality of this method and its limitations.

Multiscale Approach for Unconventional Superconductors

The remarkable electromagnetic characteristics inherent in unconventional superconductors have catalyzed the advance-ment of numerous technological innovations,spanning from energy-efficient power transmission and high-field magnets to sensitive detectors and quantum computing systems.Central to the functionality of these applications lies the superconducting characteristics,which govern pivotal phenomena including Cooper pair formation and macroscopic phase coherence,resulting in the attainment of zero electrical resistance,complete diamagnetism,and the Josephson tunneling effect.The complex phases and orders in these ma-terials significantly alter their key electronic and magnetic properties,posing challenges in elucidating the underlying physics and further enhancing their functional capabilities.The multiscale approach,representing a useful strategy for understanding materials across diverse length scales using a variety of experimental tools,can reveal intricate details in real and reciprocal spaces,facilitating cross-validation.In this brief review,we introduce the principle of the multiscale approach along with examples demonstrating its efficacy in unraveling the electronic and magnetic properties of unconventional superconductors.

On Ductile Damage Modelling of Heterogeneous Material Using Second-Order Homogenization Approach

The paper deals with the numerical modelling of ductile damage responses in heterogeneous materials using the classical second-order homogenization approach.The scale transition methodology in the multiscale framework is described.The structure at the macrolevel is discretized by the triangular C^(1) finite elements obeying nonlocal continuum theory,while the discretization of microstructural volume element at the microscale is conducted by means of the mixed type quadrilateral finite element with the nonlocal equivalent plastic strain as an additional nodal variable.The ductile damage evolution at the microlevel is modelled by using the gradient enhanced elastoplasticity.The macrolevel softening is governed by two criterions expressed by the increase in homogenized damage variable and the threshold of the local equivalent strain.The softening at each material point at the macrolevel is detected by the critical value of the homogenized damage,where homogenization of the damage variable is performed onlywithin softening area.Due to the nonlocal continuumtheory applied,a realistic softening behaviour is demonstrated after the damage initiation,compared to the widely used first-order homogenization approach.All algorithms derived have been embedded into the finite element code ABAQUS by means of the user subroutines and verified on the standard benchmark problems.The damage evolution at both microlevel and macrolevel has been demonstrated.

A multiscale modeling approach to adhesive contact

In order to model the adhesive contact across different length scales, a multiscale approach is developed and used to study the adhesive contact behaviors between a rigid cylinder and an elastic face-centered cubic (FCC) substrate. The approach combines an atomistic treatment of the interfacial region with an elastic mechanics method description of the continuum region. The two regions are connected by a coupling region where nodes of the continuum region are refined to atoms of the atomistic region. Moreover, the elastic constants of FCC crystals are obtained directly from the Lennard-Jones potential to describe the elastic response characteristics of the continuum region, which ensures the consistency of material proprieties between atomistic and con- tinuum regions. The multiscale approach is examined by comparing it with the pure MD simulation, and the results indicate that the multiscale modeling approach agrees well with the MD method in studying the adhesive contact behaviors.

Multi-scale Modeling of the Ionic Diffusivity of Cement-based Materials

A new multiscale numerical approach was presented to predict the ionic diffusivity of cement based materials,which incorporated the lattice Boltzmann method,the conjugate gradient method,and the random walk method.In particular,the lattice Boltzmann method was applied to model the ionic diffusion in pore space of cement paste,while the upscaling of effective ionic diffusivity from cement paste（mortar） to concrete was processed by means of the conjugate gradient method and the random walk method.A case study was then presented,i e,the chloride diffusivity of concrete affected by sand content and gravel content.It is shown that the results of numerical prediction agree well with those of experimental measurements adopted from literatures.The multiscale numerical approach provides a prior assessment of ionic diffusivity for cement based materials from a microstructural basis.

Bridging gap between agro-industrial waste, biodiversity and mycelium-based biocomposites: Understanding their properties by multiscale methodology

A multiscale methodology approach was employed integrating microscopic analysis of the biomasses present in the biocomposite (lignocellulosic and fungal) to understand their macro- scopic response in terms of physical and mechanical properties. Colombian native strain Gano- derma gibbosum , used for the first time in the production of biocomposites was cultivated on peach palm fruit peel flour and sugar cane bagasse wet dust, individually and as a mixture. During the solid-state fermentation were monitoring the change that occurred in substrate composition such as glucan, arabinoxylan, and lignin through biomass compositional analysis using structural car- bohydrates and lignin. Moreover, fungal biomass formation was monitored via scanning electron microscopy. The resulting biocomposites underwent characterization through flexural and water absorption tests. Our findings indicated that G. gibbosum primarily degraded the polysaccharides in each of the evaluated media. However, lignin degradation to 15.06 g/g was only observed in the mixture biocomposite of peach palm fruit peel fluor and sugarcane bagasse wet dust in a ratio of 1 ꞉1, accompanied by a reduction in glucan and arabinoxylan weights to 26.1 and 7.72 g/g, respectively. This polymer degradation, combined with a protein-rich source in the mix- ture biocomposite of peach palm fruit peel fluor and sugarcane bagasse wet dust in a ratio of 1 ꞉1, facilitated the production of a fungal skin (biological matrix) with a high hyphal density of 65%, contributing to Young’s modulus of 3.83 MPa, elongation without failure, and low wa- ter absorption rate in this biocomposite (55%). The lignocellulosic biomass in the culture media acted as a filler for mechanical interlocking with the matrix and provided attachment points for water absorption. Thus, our study establishes a connection between the microscopic scale and the macroscopic behavior of this biocomposite, assessing structural carbohydrates and lignin analy- sis during solid-state fermentation (SSF), laying the groundwork for a more customized design of mycelium-based biocomposites. Finally, this study demonstrates the possibility of tailoring nutrient composition by designing their culture media to obtain physical-mechanical properties according to the application requirement.

A MULTISCALE MODELING APPROACH INCORPORATING ARIMA AND ANNS FOR FINANCIAL MARKET VOLATILITY FORECASTING

The financial market volatility forecasting is regarded as a challenging task because of irreg ularity, high fluctuation, and noise. In this study, a multiscale ensemble forecasting model is proposed. The original financial series are decomposed firstly different scale components （i.e., approximation and details） using the maximum overlap discrete wavelet transform （MODWT）. The approximation is pre- dicted by a hybrid forecasting model that combines autoregressive integrated moving average （ARIMA） with feedforward neural network （FNN）. ARIMA model is used to generate a linear forecast, and then FNN is developed as a tool for nonlinear pattern recognition to correct the estimation error in ARIMA forecast. Moreover, details are predicted by Elman neural networks. Three weekly exchange rates data are collected to establish and validate the forecasting model. Empirical results demonstrate consistent better performance of the proposed approach.