Elucidating the constitutive relationship of calcium-silicate-hydrate gel using high throughput reactive molecular simulations and machine learning.

Prediction of material behavior using machine learning (ML) requires consistent, accurate, and, representative large data for training. However, such consistent and reliable experimental datasets are not always available for materials. To address this challenge, we synergistically integrate ML with high-throughput reactive molecular dynamics (MD) simulations to elucidate the constitutive relationship of calcium-silicate-hydrate (C-S-H) gel-the primary binding phase in concrete formed via the hydration of ordinary portland cement. Specifically, a highly consistent dataset on the nine elastic constants of more than 300 compositions of C-S-H gel is developed using high-throughput reactive simulations. From a comparative analysis of various ML algorithms including neural networks (NN) and Gaussian process (GP), we observe that NN provides excellent predictions. To interpret the predicted results from NN, we employ SHapley Additive exPlanations (SHAP), which reveals that the influence of silicate network on all the elastic constants of C-S-H is significantly higher than that of water and CaO content. Additionally, the water content is found to have a more prominent influence on the shear components than the normal components along the direction of the interlayer spaces within C-S-H. This result suggests that the in-plane elastic response is controlled by water molecules whereas the transverse response is mainly governed by the silicate network. Overall, by seamlessly integrating MD simulations with ML, this paper can be used as a starting point toward accelerated optimization of C-S-H nanostructures to design efficient cementitious binders with targeted properties.

Health Degradation Monitoring and Early Fault Diagnosis of a Rolling Bearing Based on CEEMDAN and Improved MMSE.

Rolling bearings play a crucial role in rotary machinery systems, and their operating state affects the entire mechanical system. In most cases, the fault of a rolling bearing can only be identified when it has developed to a certain degree. At that moment, there is already not much time for maintenance, and could cause serious damage to the entire mechanical system. This paper proposes a novel approach to health degradation monitoring and early fault diagnosis of rolling bearings based on a complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and improved multivariate multiscale sample entropy (MMSE). The smoothed coarse graining process was proposed to improve the conventional MMSE. Numerical simulation results indicate that CEEMDAN can alleviate the mode mixing problem and enable accurate intrinsic mode functions (IMFs), and improved MMSE can reflect intrinsic dynamic characteristics of the rolling bearing more accurately. During application studies, rolling bearing signals are decomposed by CEEMDAN to obtain IMFs. Then improved MMSE values of effective IMFs are computed to accomplish health degradation monitoring of rolling bearings, aiming at identifying the early weak fault phase. Afterwards, CEEMDAN is performed to extract the fault characteristic frequency during the early weak fault phase. The experimental results indicate the proposed method can obtain a better performance than other techniques in objective analysis, which demonstrates the effectiveness of the proposed method in practical application. The theoretical derivations, numerical simulations, and application studies all confirmed that the proposed health degradation monitoring and early fault diagnosis approach is promising in the field of prognostic and fault diagnosis of rolling bearings.