Mehdi Shishehbor

Anomalies are samples that significantly deviate from the rest of the data and their detection plays a major role in building machine learning models that can be reliably used in applications such as data-driven design and novelty detection. The majority of existing anomaly detection methods either are exclusively developed for (semi) supervised settings, or provide poor performance in unsupervised applications where there are no training data with labeled anomalous samples. To bridge this research gap, we introduce a robust, efficient, and interpretable methodology based on nonlinear manifold learning to detect anomalies in unsupervised settings. The essence of our approach is to learn a low-dimensional and interpretable latent representation (aka manifold) for all the data points such that normal samples are automatically clustered together and hence can be easily and robustly identified. We learn this low-dimensional manifold by designing a learning algorithm that leverages either a latent map Gaussian process (LMGP) or a deep autoencoder (AE). Our LMGP-based approach, in particular, provides a probabilistic perspective on the learning task and is ideal for high-dimensional applications with scarce data. We demonstrate the superior performance of our approach over existing technologies via multiple analytic examples and real-world datasets.

Bayesian optimization (BO) is a sequential optimization strategy that is increasingly employed in a wide range of areas such as materials design. In real-world applications, acquiring high-fidelity (HF) data through physical experiments or HF simulations is the major cost component of BO. To alleviate this bottleneck, multi-fidelity (MF) methods are used to forgo the sole reliance on the expensive HF data and reduce the sampling costs by querying inexpensive low-fidelity (LF) sources whose data are correlated with HF samples. However, existing multi-fidelity BO (MFBO) methods operate under the following two assumptions that rarely hold in practical applications: (1) LF sources provide data that are well correlated with the HF data on a global scale, and (2) a single random process can model the noise in the MF data. These assumptions dramatically reduce the performance of MFBO when LF sources are only locally correlated with the HF source or when the noise variance varies across the data sources. In this paper, we view these two limitations and uncertainty sources and address them by building an emulator that more accurately quantifies uncertainties. Specifically, our emulator (1) learns a separate noise model for each data source, and (2) leverages strictly proper scoring rules in regularizing itself. We illustrate the performance of our method through analytical examples and engineering problems in materials design. The comparative studies indicate that our MFBO method outperforms existing technologies, provides interpretable results, and can leverage LF sources which are only locally correlated with the HF source.

The interfacial adhesion between asphalt binder and carbon nanotubes (CNTs) depends on many nanoscopic properties such as diffusion of SARA molecules on CNTs surface. Functionalization of CNTs with Oxygens (O=CNTs), hydroxyl groups (HO–CNTs), and hydrogens (H–CNTs) has been an effective way to modify the surface properties of CNTs and ultimately the macroscopic properties of the CNT-composites. This paper presents the effect of different dosages of oxygenated and hydrogenated CNTs on the adhesion and diffusion of SARA molecules on CNTs’ surfaces. First, reactive molecular dynamics simulation is used to oxygenate and hydrogenate CNTs up to a certain dosage. Next, it is employed to model the interaction and diffusion of SARA molecules with the functionalized CNTs. We employ the steer molecular dynamic (SMD) and Einstein formula to calculate the adhesion and diffusion properties. The results demonstrate that hydrogenation has little effect on the adhesion energy, while oxygenation can increase adhesion energy up to 100% for 25% dosage. The diffusion coefficient dramatically drops for both oxygenated and hydrogenated CNTs, with lower values for the latter. We observe that for hydrogenated and oxygenated CNTs at different dosages, asphaltene, resin, aromatic, and saturate molecules have the highest to lowest values, respectively.

Nature utilizes the available resources to construct lightweight, strong and tough materials under constrained environmental conditions. The impact surface of the fast-striking dactyl club from the mantis shrimp is an example of one such composite material; the shrimp has evolved the capability to localize damage and avoid catastrophic failure from high-speed collisions during its feeding activities. Here we report that the dactyl club of mantis shrimps contains an impact-resistant coating composed of densely packed (about 88 per cent by volume) ~65-nm bicontinuous nanoparticles of hydroxyapatite integrated within an organic matrix. These mesocrystalline hydroxyapatite nanoparticles are assembled from small, highly aligned nanocrystals. Under impacts of high strain rates (around 10⁴ s⁻¹), particles rotate and translate, whereas the nanocrystalline networks fracture at low-angle grain boundaries, form dislocations and undergo amorphization. The interpenetrating organic network provides additional toughening, as well as substantial damping, with a loss coefficient of around 0.02. An unusual combination of stiffness and damping is therefore achieved, outperforming many engineered materials.

Improving the adhesion properties of carbon nanotubes (CNTs) at the molecular scale can significantly enhance dispersion of CNT fibers in polymer matrix and unleash the dormant extraordinary mechanical properties of CNTs in CNT-polymer nanocomposites. Inspired by the outstanding adhesion, dispersion, mechanical, and surface functionalization properties of crystalline nanocellulose (CNC), this paper studies the mechanical and adhesion properties of CNT wrapped by aligned cellulose chains around CNT using molecular dynamic simulations. The strength, elastic modulus, and toughness of CNT-cellulose fiber for different cellulose contents are obtained from tensile and compression tests. Additionally, the effect of adding cellulose on the surface energy, interfacial shear modulus, and strength is evaluated. The result shows that even adding a single layer cellulose wrap (≈55% content) significantly decreases the mechanical properties, however, it also dramatically enhances the adhesion energy, interfacial shear strength, and modulus. Adding more cellulose layers, subsequently, deceases and increases mechanical properties and adhesion properties, respectively. In addition, analysis of nanopapers of pristine CNT, pristine CNC, and CNT-wrapped cellulose reveals that CNT-wrapped cellulose nanopapers are strong, stiff, and tough, while for CNT and CNC either strength or toughness is compromised. This research shows that cellulose wraps provide CNT fibers with tunable mechanical properties and adhesion energy that could yield strong and tough materials due to the excellent mechanical properties of CNT and active surface and hydrogen bonding of cellulose.

During recent decade, the pavement sustainability has received much attention by road agencies, companies, governments and research institutes. The aim of this paper is to introduce and evaluate the sustainability of the technologies developed or underdeveloped to address environmental issues of asphalt mixtures, and the waste materials and by-products that can be recycled and reused in asphalt production. Warm Mix Asphalt (WMA) technology, the most popular waste materials to substitute neat binder (crumb rubber modifier (CRM), recycled plastic (RP), vacuum tower bottoms (VTB)) and/or virgin aggregates (reclaimed asphalt pavement (RAP), reclaimed asphalt shingles (RAs), construction and demolition (C and D) wastes, steel and copper slags), and bio-binders were evaluated with respect to their environmental and economic benefits and engineering performance as the main components of pavement sustainability. The performance evaluation was carried out by examining rutting, moisture susceptibility, thermal and fatigue cracking resistance. Two main environmental impacts, greenhouse gas (GHG) emission, and energy consumption were considered to study the environmental effects of these technologies and materials. Additionally, the economic effect was investigated considering initial cost and long-term benefit. Results of investigation illustrated that although each individual technology and waste material is valuable in terms of performance and/or the environment, specific combinations such as WMA + RAP, WMA + CRM, RAP + CRM, and WMA + CRM + RAP lead to further benefits. Notably, these combinations suffer from a lack of comprehensive economic analysis, thus, their sustainability cannot be fully assessed and is prone to future studies.