Search

Electrochemically converting CO2 to value-added multi-carbon (C2+) fuels and chemicals is a favorable way to achieve carbon neutrality. Herein, polyaniline/CuO nanosheets (PANI/CuO NSs) hybrid electrocatalysts are developed in order to achieve superior C2+ selectivity by imparting PANI functional component to the CuO NSs. The decorated PANI nanoparticles (NPs) can effectively stabilize the *CO intermediates and increase their coverage on the active Cu sites, which facilitates the C–C coupling to form multi-carbon products. Benefiting from the synergetic effect of PANI and CuO NSs, best Faradaic efficiency (FE) for C2+ product up to 66.4% at −1.6 V vs. reversible hydrogen electrode (RHE) in a H-cell measurement and 60.0% at 400 mA·cm−2 in a flow cell measurement are demonstrated by PANI/CuO NSs-25 sample. More importantly, the C2+ selectivity keeps stable even in a continuous measurement time period of 92 h in H-cell measurement. The present study may provide more insights for designing efficient hybrid materials toward superior C2+ production from electrocatalytic CO2 reduction. [ABSTRACT FROM AUTHOR]

This paper discusses back-propagation neural networks (NN) for predicting the settlement during tunneling. Three settlement parameters and 11 major affecting factors have been identified from analyzing the general tunneling operations. A general neural network model is trained and tested using the actual collected data from the 6.5 km Brasilia Tunnel in Brazil. The general model generates an average error of 70 mm for the predicted settlements compared with the actual values. To improve the prediction accuracy, modular NN models are studied based on the concept of integrating multiple NN modules in one system with each module being constrained to operate at one specific situation of a complicated real world problem. The modular concept can make better use of neural computation algorithms to improve the convergence in the training process. It has been studied on modeling multiple output variables and discrete input variables. After applying modular models to the same Brasilia Tunnel, the average prediction error is reduced to 33.4 mm, which shows a significant improvement over the general NN model.