Search

As energy conversion systems continue to grow in complexity, pneumatic control valves may exhibit unexpected anomalies or trigger system shutdowns, leading to a decrease in system reliability. Consequently, the analysis of time-domain signals and the utilization of artificial intelligence, including deep learning methods, have emerged as pivotal approaches for addressing these challenges. Although deep learning is widely used for pneumatic valve fault diagnosis, the success of most deep learning methods depends on a large amount of labeled training data, which is often difficult to obtain. To address this problem, a novel fault diagnosis method based on the attention-weighted relation network (AWRN) is proposed to achieve fault detection and classification with small sample data. In the proposed method, fault diagnosis is performed through the relation network in few-shot learning, and in order to enhance the representativeness of feature extraction, the attention-weighted mechanism is introduced into the relation network. Finally, in order to verify the effectiveness of the method, a DA valve fault dataset is constructed, and experimental validation is performed on this dataset and another benchmark PU rolling bearing fault dataset. The results show that the accuracy of the network on DA is 99.15%, and the average accuracy on PU is 98.37%. Compared with the state-of-the-art diagnosis methods, the proposed method achieves higher accuracy while significantly reducing the amount of training data.

A bifunctional catalyst comprising CuCl 2 /Al 2 O 3 and nitrogen-doped carbon was developed for an efficient one-pot ethylene oxychlorination process to produce vinyl chloride monomer (VCM) up to 76 % yield at 250 °C and under ambient pressure, which is higher than the conventional industrial two-step process (≈50 %) in a single pass. In the second bed, active sites containing N-functional groups on the metal-free N-doped carbon catalyzed both ethylene oxychlorination and ethylene dichloride (EDC) dehydrochlorination under the mild conditions. Benefitting from the bifunctionality of the N-doped carbon, VCM formation was intensified by the surface Cl*-looping of EDC dehydrochlorination and ethylene oxychlorination. Both reactions were enhanced by in situ consumption of surface Cl* by oxychlorination, in which Cl* was generated by EDC dehydrochlorination. This work offers a promising alternative pathway to VCM production via ethylene oxychlorination at mild conditions through a single pass reactor., (© 2020 The Authors. Published by Wiley-VCH GmbH.)

The understanding of the water effect on olefin selectivity in Fischer-Tropsch synthesis (FTS) is limited by the complexity of the reaction network. Herein, we employ propene hydrogenation as a model reaction to isolate the water effect on olefin adsorption and hydrogenation from the complex reaction of FTS. It is clearly observed that the added water inhibits the activity of propene hydrogenation on two cobalt catalysts supported on high-surface-area alumina (HAS Al 2 O 3 ) and low-surface-area alumina (LSA Al 2 O 3 ), respectively. The inhibiting effect is much stronger for Co/HSA Al 2 O 3 . DFT investigation demonstrates that the in situ generated OH, rather than H 2 O and O, impedes the adsorption of propene and thus decreases the activity of propene hydrogenation. The suppressive effect of OH on propene adsorption is attributed to the downshift of the d-band center and the Bader charge of the catalyst surface. The DFT-based kinetic analysis finds that the higher site coverage of OH results in the more pronounced negative effect on propene hydrogenation. Furthermore, the theory of OH-induced weak olefin adsorption and low olefin hydrogenation activity could rationalize the enhancement effect of water on the olefin selectivity and the particle size dependence of the water effect in FTS. The insights obtained here may inspire researchers to optimize olefin selectivity by manipulating the electronic properties of catalysts with hydroxyl species.

Understanding the scaling relations of adsorption energies and activation energies greatly facilitates the computational catalyst design. To reduce the computational cost and guarantee efficiency, improved scaling relations were advocated in this study to rapidly acquire the energetics for transition metal surface reactions and further to rapidly and effectively map the activity of transition-metal catalysts. The overall catalytic activity for the surface reactions between C-, H- and O-containing species could be related to their adsorption energies using C, H and O binding energies as descriptors via improved scaling relations. The UBI-QEP (unity bond index-quadratic exponential potential) method, one of the scaling relations used to estimate the adsorption energies from descriptors, was significantly improved by taking into account the changes in the A-B bond indexes during adsorption and the molecular structure of adsorbed species using density functional theory (DFT) data as a benchmark. The improved UBI-QEP approach could satisfactorily predict the DFT (BEEF-vdW) and experimental adsorption energies. DFT calculations with the BEEF-vdW functional were also employed for establishing the BEP (Brønsted-Evans-Polanyi) relationships as scaling relations to correlate the reaction heats with activation energies for C-H, C-O, C-C, and O-H bond cleavages and recombination. The capability of the improved UBI-QEP+BEP approach was tested as a generic framework to map the activity trend for steam methane reforming (a probe reaction) through microkinetic modeling. The results demonstrated that our approach reduces the computational cost by six orders of magnitude while maintaining a reasonable degree of accuracy as compared to the DFT (BEEF-vdW) and experimental approaches.

Potassium (K), an important impurity in syngas from biomass, can have a large influence on the activity and selectivity of cobalt-based Fischer-Tropsch synthesis (FTS) catalysts in Biomass to Liquids (BTL) processes. In this work, the potassium adsorption behavior on hcp cobalt was systematically studied using density functional theory. The surface energy calculations and Wulff construction of the equilibrium shape of hcp cobalt showed it is dominated by 10 facets. The interaction of K with these facets has been investigated. The results showed that the stepped facet (10-12) has the highest K adsorption energy of -2.40 eV. The facets (0001), (10-10), (10-11), (10-15), and (21-30) also showed relatively high K adsorption energies in the range of -2.28 to -2.34 eV. The corrugated facets exhibited comparatively lower K adsorption energies (-2.04 to -2.18 eV), and would be less favorable for K adsorption. It was also found that the adsorption properties depend on coverage, where the K adsorption energy decreased with increasing coverage. Diffusion energy barrier calculations indicated that K was mobile on typical facets (0001) and (10-11) with very low diffusion barriers (<0.15 eV). On stepped facets, although K could move freely along the same step (diffusion barrier <0.01 eV), diffusion from one step to another had a significantly higher barrier of 0.56 eV. This suggested that K atoms would be mobile to some extent during FTS reaction conditions, and tend to occupy the most favorable sites independent of their initial position. The results obtained in this work provide valuable information on the interaction of K with cobalt surfaces, relevant for practical cobalt catalysts and their application in BTL processes.