Search

This study aims to develop multifunctionality of cementitious composites with the integrated self-sensing and self-healing capacities by incorporating conductive carbon black (CB) with CB-encapsulated slaked lime (SL). The microsized SL particles were premixed with a half of designed content of nanosized CB particles. When CB agglomerations coat around the SL surfaces, SL does not hydrate until the CB coating is removed. Another half of designed weight of CB is uniformly dispersed using ultrasonication with superplasticizer and added to obtain piezoresistivity. The results show that the stress sensing capacity of CB-SL-cementitious composite performs well with the compressive stress. Autogenous healing performances presented significantly can improve the self-healing capacity with the increase of SL. Furthermore, the healing efficiency is affected by the crack width and dispersion of SL, and the smaller cracks with SL are more easily healed. The size of CB agglomerations decreases with the added SL, and the main product of self-healing is calcium carbonate.

This paper presents coupled hydro-mechanical modelling of unsaturated soil problems by incorporating some advanced numerical and constitutive models in a general-purpose commercial software package, Abaqus. Two different strategies for the interpretation of unsaturated soil behaviour with respect to the soil volume change are considered and the relevant constitutive models are implemented through user-defined subroutines. The first approach that is commonly used in geomechanics considers suction as an additional variable and uses a constitutive model in the space of effective stress–suction assuming that soil compressibility is a function of suction. The second approach treats the constitutive model in the space of effective stress and degree of saturation assuming that the soil compressibility is a function of the degree of saturation. The hydro-mechanical behaviour regarding the change in the degree of saturation caused by both suction and net stress changes is also considered, together with the effect of hydraulic hysteresis. Validation of the implemented algorithms is presented through several benchmark problems. Finally, the application and utility of the implemented procedures are illustrated by simulations of two challenging problems of unsaturated geomechanics, including a slope failure due to seepage and rainfall infiltration as well as cone penetration tests in unsaturated soil. A suitable mesh optimisation scheme is also incorporated to handle finite deformations.

When soil particles break, the particle size distribution (PSD) becomes a variable in the same way as other variables like void ratio, soil structure and anisotropy, etc. To consider particle breakage in a constitutive model, we need three key components: (i) quantification of PSD in a realistically simple manner, (ii) evolution of PSD during particle breakage, and (iii) influences of PSD on other soil properties like strength and stiffness. This paper firstly summarizes the latest advances in the first two components, discusses new ways of quantifying PSD effects, and finally presents a new critical state model where the PSD is treated as a variable. In discussing the PSD effects, we focus on the movement of the critical state line (CSL) due to particle breakage. We introduce a new state parameter and a new evolution law of the CSL. We assume that the CSL shifts downwards in the v—lnp space with increasing particle breakage under relatively low stresses, but all the CSLs for different PSDs converge to a steady state at high stresses where particle breakage eventually stops and is no longer the main mechanism for soil deformation. The proposed model is compared with other constitutive model in the literature and validated against experimental data, which demonstrates its satisfactory performance.

Particle breakage of carbonate sands is widely encountered during the construction of harbor facilities. In this study, a series of one-dimensional compression tests on carbonate sands with different initial particle size distributions (PSDs) and initial void ratios were completed at high vertical stress (32 MPa), to investigate the influence of PSD and void ratio on the compression behavior of carbonate sands and the corresponding breakage properties. The PSDs used in this study were prepared by fractal distribution with two different fractal dimensions, i.e., 0.5 and 2.0. The results show that samples with fractal dimension of 0.5, where the initial PSDs are poorly graded, have a unique normal compression line (NCL), implying the occurrence of non-transitional behavior. However, when the PSD of samples tend to be better graded with fractal dimension of 2.0, non-convergent compression paths are likely to occur, which means a transitional behavior is identified. Particle breakage is observed after compression tests on samples with two different initial PSDs. It can, therefore, be concluded that particle breakage may have happened in both transitional and non-transitional behavior soils.

Rainfall-induced landslides represent a severe hazard around the world due to their sudden occurrence, as well as their widespread influence and runout distance. Considering the spatial variability of soil, stochastic analysis is often conducted to give a probability description of the runout. However, rainfall-induced landslides are complex and time-consuming for brute-force Monte Carlo analyses. Therefore, new methods are required to improve the efficiency of stochastic analysis. This paper presents a framework to investigate the influence and runout distance of rainfall-induced landslides with a two-step simulation approach. The complete process, from the initialization of instability to the post-failure flow, is simulated. The rainfall infiltration process and initialization of instability are first solved with a coupled hydro-mechanical finite element model. The post-failure flow is simulated using the coupled Eulerian–Lagrangian method, wherein the soil can flow freely in fixed Eulerian meshes. An equivalent-strength method is used to connect two steps by considering the effective stress of unsaturated soil. A rigorous method has been developed to accurately quantify the influence and runout distance via Eulerian analyses. Several simulations have been produced, using three-dimensional analyses to study the shapes of slopes and using stochastic analysis to consider uncertainty and the spatial variability of soils. It was found that a two-dimensional analysis assuming plain strain is generally conservative and safe in design, but care must be taken to interpret 2D results when the slope is convex in the longitudinal direction. The uncertainty and spatial variability of soils can lead to the statistic of influence and runout distance. The framework of using machine-learning models as surrogate models is effective in stochastic analysis of this problem and can greatly reduce computational effort.

A novel cement-based sensors was developed with integrated self-sensing superhydrophobicity, and self-cleaning functions in this paper. The synthesis was carried out by penetrating precast graphene nanoplate/cement-based sensors with silane/isopropanol solutions. The silane-treated cement-based sensors showed satisfactory stress/strain sensing performance with an average gauge factor of 141.8, and exhibited excellent hydrophobic behaviour with the highest water contact angle of 163° on the intact surface. The contact angle decreased to 148° and 142°, for the surface with scratches and for the inner part of sensors, respectively. The reduction was due to the spalling and less amount of silane particles within the scratches and the harder entry of silane to the inner part of sensor. The self-cleaning properties of silane-treated cement-based sensor were evaluated by the visual observation of removing efficiency of hydrophilic carbon black dust and lipophilic sauces after water rinsing. It was found that the silane-treated cement-based sensor showed excellent self-cleaning performance using hydrophilic carbon dust. Despite the removing efficiency decreased for the lipophilic sauces, the silane-treated cement-based sensors maintained much less stain than that of untreated ones on the surface. The related results will promote the synthesis and practical applications of multifunctional cement-based sensors for the application of intrisic structural health monitoring.

Using machine-learning models as surrogate models is a popular technique to increase the computational efficiency of stochastic analysis. In this technique, a smaller number of numerical simulations are conducted for a case, and obtained results are used to train machine-learning surrogate models specific for this case. This study presents a new framework using deep learning, where models are trained with a big dataset covering any soil properties, spatial variabilities, or load conditions encountered in practice. These models are very accurate for new data without re-training. So, the small number of numerical simulations and training process are not needed anymore, which further increases efficiency. The prediction of bearing capacity of shallow strip footings is taken as an example. We start with a simple scenario, and progressively consider more complex scenarios until the full problem is considered. More than 12,000 data are used in training. It is shown that one-hidden-layer fully connected networks can give reasonable results for simple problems, but they are ineffective for complex problems, where deep neural networks show a competitive edge, and a deep-learning model achieves a very high accuracy (the root-mean-square relative error is 3.1% for unseen data). In testing examples, this model is proven very accurate if the parameters of specific cases are well in the defined limits. Otherwise, the capability of deep-learning models can be extended by simply generating more data outside the current limits and re-training the models.

Pore ice content is crucial in evaluating the mechanical properties of frozen soils. Existing strength criterion models are usually empirical and ignore the influence of pore ice content. By assuming that the critical state shear stress ratio of soil is a function of the stress level, a critical state line of frozen soil is proposed to consider pore ice content. By combining the Mohr-Coulomb (M-C) and Drucker-Prager strength criteria to describe the failure shape characteristics on the deviatoric plane, a new strength criterion is established for complex stress conditions. The proposed model is validated against existing models and measured data in the literature. In addition, the proposed model can uniformly describe the CSL of different types of geotechnical materials and has a clear physical meaning, which may provide a theoretical basis for constitutive models.

Purpose: Prediction of excess pore water pressure and estimation of soil parameters are the two key interests for consolidation problems, which can be mathematically quantified by a set of partial differential equations (PDEs). Generally, there are challenges in solving these two issues using traditional numerical algorithms, while the conventional data-driven methods require massive data sets for training and exhibit negative generalization potential. This paper aims to employ the physics-informed neural networks (PINNs) for solving both the forward and inverse problems. Design/methodology/approach: A typical consolidation problem with continuous drainage boundary conditions is firstly considered. The PINNs, analytical, and finite difference method (FDM) solutions are compared for the forward problem, and the estimation of the interface parameters involved in the problem is discussed for the inverse problem. Furthermore, the authors also explore the effects of hyperparameters and noisy data on the performance of forward and inverse problems, respectively. Finally, the PINNs method is applied to the more complex consolidation problems. Findings: The overall results indicate the excellent performance of the PINNs method in solving consolidation problems with various drainage conditions. The PINNs can provide new ideas with a broad application prospect to solve PDEs in the field of geotechnical engineering, and also exhibit a certain degree of noise resistance for estimating the soil parameters. Originality/value: This study presents the potential application of PINNs for the consolidation of soils. Such a machine learning algorithm helps to obtain remarkably accurate solutions and reliable parameter estimations with fewer and average-quality data, which is beneficial in engineering practice.

Data-driven intelligent surrogate models gain popularity recently. Particularly in Monte-Carlo-style stochastic analysis, the influencing factors are considered as inputs, the quantities of interest are considered as outputs, and cheaper-to-evaluate surrogates models are built from a small amount of sample data and are used for the full Monte-Carlo analysis. This paper presents a framework with three innovations: (1) we build surrogate models for a particular problem that covers any possible material properties or boundary conditions commonly encountered in practice, so the models are ready to use, and do not require new data or training anymore. (2) The inputs and outputs to the problem are both spatially variable. Even after discretization, the input and output sizes are in the order of tens of thousands, which is challenging for traditional machine-learning algorithms. We take the footing failure mechanism as an example. Two types of neural networks are examined, fully connected networks and deep neural networks with complicated non-sequential structures (a modified U-Net). (3) This study is also the first attempt to use U-Nets as surrogate models for geotechnical problems. Results show that fully connected networks can fit well simple problems with a small input and output size, but fail for complex problems. Deep neural networks that account for the data structure give better results.

Carbonate soils are high in compressibility and low in particle strength, and therefore need special treatment for many geotechnical applications. This paper presents comprehensive one-dimensional compression tests to investigate the compression behavior and particle breakage of carbonate soils with various initial particle size distributions (PSDs) and a wide range of initial void ratios at a high effective vertical stress of 32 MPa. A clear shape transition in terms of the void ratio against effective vertical stress was observed with varying silt fraction and void ratio. The deformation mechanism for the samples with less silt fraction and higher void ratio is mainly dominated by the particle-breakage-induced unlocking effect. Samples with more silt fraction and lower void ratio will experience less particle breakage and their deformation mechanism is mainly attributed to interparticle locking effect. A ‘transitional’ mode of behavior without a unique normal compression line was identified for the samples with silt fraction ranging from 1.6% to 33%. It was also found that carbonate soil samples with both ‘transitional’ and ‘non-transitional’ modes of compression behavior may both experience considerable amount of particle breakage and show different trends in the evolution of particle shape.

This paper presents experimental investigation into the effects of particle size distribution of subgrade soil on mud pumping. The results show that subgrade soils with higher fine contents do not necessarily lead to more serious mud pumping. A soil with a higher silt content tends to cause the formation of a less permeable interlayer at the bottom of the ballast, which effectively reduces the particle migration magnitude. Increasing the median particle size (d50) or reducing the coefficient of uniformity (d60/d10) of the studied sandy silt promotes the migration distance of particles. While mud pumping is essentially an internal erosion problem caused by cyclic loads, existing filter theories do not directly apply to mud pumping. The findings from this study can help selecting proper rail embankment fills to reduce mud pumping.

Tunnel Boring Machines (TBMs) have been increasingly used in tunnelling projects. Forecasting future TBM performance would be desirable for project time management and cost control. We aim to use recurrent neural networks to predict the near future TBM penetration rate from historical data. Our datasets are composed of Changsha and Zhengzhou metro lines, with totally different geological conditions. In the experiments, the one-step forecast of TBM penetration rate by the traditional recurrent neural network (RNN) or long short-term memory (LSTM) is relatively accurate, irrespective of the different geological conditions used in training and evaluation. Predicting the next Nth step penetration rate proves to be more challenging and depends on the time to the future or the distance ahead of the TBM cutterhead. There are generally time lags between measured and predicted results. The recursive RNN is then developed to address the lag problems, but to no avail. Alternative methods for predicting future penetration rates are studied, including the penetration rate at the Nth step in the future and the average penetration rate of the next N steps, with the latter being trained by long-input or short-input methods. The average N-step forecast using short inputs provides the best results, and its performance over other alternatives becomes more distinct as the number N increases. We also discuss the possibility of the forecast problem as a quasi-random walk, which means that forecasting penetration rate cannot easily be achieved using low-frequency data with RNNs, and that the accuracy depends on the correlation between the last and predicted steps in the data.

Silicone hydrophobic powder (SHP) and crystalline waterproofing admixture (CWA) were used to improve the impermeability of carbon black (CB)/cement-based sensors. The mechanical, electrical and piezoresistive properties, waterproofing and chloride resistance of CB/cementitious composites were investigated in this study. The piezoresistivity before or after different durations of immersion in freshwater and 3% sodium chloride solution and the stability in freshwater and marine environment were studied and compared. The results show that compressive strength increased with the additions of CWA and SHP, while the tensile strength slightly decreased with CWA, due to the formation of crystalline. Moreover, cementitious composites with SHP exhibited the best water impermeability, while the counterpart containing CWA presented the optimal chloride resistance. Although cementitious composites with SHP exhibited the highest electrical resistivity, the most stable piezoresistivity occurred after 90 days of immersion in freshwater. On the other hand, cementitious composites incorporating CWA presented the lowest electrical resistivity, but the piezoresistivity continually decreased with the immersion duration. Because of the free ions, piezoresistivity increased as a result of the immersion in sodium chloride solution. The related results will provide an insight into the piezoresistivity of hydrophobic cement-based sensors under moisture and chloride environments for future structural health monitoring.

Heterogeneity of interfacial transition zones (ITZs) is a key factor for the properties and failure mechanism of geopolymer concrete. The nano/microscale properties and heterogeneity of the ITZs (the top, bottom and lateral interfaces) prepared by encompassing polished aggregates in the modelled fly ash-based geopolymer concrete were statistically investigated in this study. The nanoindentation and nanoscratch results show that the nano/micromechanical properties of the gel-related phases of ITZs at the top and bottom boundaries are higher than the corresponding ones at the lateral boundaries and bulk paste. The mechanism of the better properties of ITZs at the top and bottom boundaries is unveiled based on quantitative image analysis of the amount, diameter and proportion distribution of fly ash particles. A strategy of controlling heterogeneity of ITZs and using polished aggregates, rapid scratch and statistical analysis is proposed to investigate more complicated ITZs within acceptable testing duration.

The electrical, mechanical properties, and piezoresistive performances of intrinsic graphene nanoplate (GNP)/cementitious composites were investigated after subjected to impact load in this paper. The stabilized electrical resistivity before/after exposure to impact load and real-time electrical response under dynamic load were simultaneously studied. The cement hydration and microstructures of (GNP)/cementitious composites were characterized by thermal gravity analysis (TGA) and scanning electron microscope. The nearly identical hydration degree of 1.0% GNP filled cement mortar (1GNPCM) and mortar with 2% GNP (2GNPCM) indicates the physical interactions between the GNP and cement matrix. The excellent intrinsic physical properties of GNP played an important role in the enhancements of GNP/cementitious composites. After exposed to impact, the stabilized electrical resistivity, mechanical performance, and piezoresistivity of 1GNPCM were greatly changed, whereas the counterpart of 2GNPCM was well-maintained and nearly unaffected. Therefore, the severe microstructural deteriorations in 1GNPCM could be responsible for the variations, which damaged the conductive passages. The almost unchanged mechanical, electrical and piezoresistive properties enable 2GNPCM as a promising cement-based senor to provide stable piezoresistivity even after exposure to impact load. The related outcomes provide an insight into the development of impact-resistant cement-based sensors and promote the applications of cement-based sensors under extreme loading conditions.

A first order accurate fully implicit integration scheme and four different order explicit substepping integration schemes with automatic error control are used in this paper to integrate the constitutive relations of a critical state model for saturated soils. Their respective computational performance in terms of accuracy and efficiency is assessed in order to provide practical guidance for deciding which of the five is most suitable for solving numerical problems in geotechnical engineering involving critical state models. Even though existing literature of integration schemes applied to geotechnical problems has traditionally been focussed on the first order accurate implicit backward Euler and on the second order accurate explicit modified Euler with substepping almost exclusively, the findings of this paper suggest that the little extra work required in the implementation of an explicit third order Runge-Kutta substepping scheme is worth the effort, especially in terms of computational cost.