Corrosion inhibitors are one of the most effective anticorrosion techniques in reinforced concrete structures. Molecule dynamics (MD) was usually utilized to simulate the interaction between the inhibitor molecules and the surface of Fe to evaluate the inhibition effect, ignoring the influence of cement hydration products. In this paper, the adsorption characteristics of five types of common alkanol-amine inhibitors on C-S-H gel in the alkaline liquid environment were simulated via the MD and the grand canonical Monte Carlo (GCMC) methods. It is found that, in the MD system, the liquid phase environment had a certain impact on the adsorption configuration of compounds. According to the analysis of the energy, the binding ability of MEA on the surface of the C-S-H gel was the strongest. In the GCMC system, the adsorption of MEA was the largest at the same temperature. Furthermore, for the competitive adsorption in the GCMC system, the adsorption characteristics of the inhibitors on the C-S-H gel were to follow the order: MEA>DEA>TEA>NDE>DETA. Both MD and GCMC simulations confirmed that the C-S-H gel would adsorb the organic inhibitors to a different extent, which might have a considerable influence on the organic inhibitors to exert their inhibition effects.

No abstract

The solidification of network parameter values for most relation extraction models after training makes the model overconfident in the process of prediction and classification tasks. Moreover, the interference of similar relation in the text data will affect the effect of relation extraction. We propose a relation extraction model based on relation similarity and Bayesian neural network. This model uses the logistic regression loss function to make the training parameters closer to the target relation and away from the similar relation, thereby eliminating the interference of the similar relation in the relation extraction task. In addition, it is also possible to learn a probability distribution from the weights of the Bayesian-LSTM (Bayesian-Long Short-Term Memory) neural network, and retain the corresponding uncertainty on the basis of obtaining long-distance dependence information using LSTM, so that the model learns more data features while performing regularization at the weight level. The model also uses an attention mechanism to pay more attention to useful information. The experimental results on the Wikipedia data set TACRED (TAC Relation Extraction Dataset) data set show that the proposed method effectively improves the effect of the entity relation extraction model.

No abstract