Authentic surface structures under reaction conditions determine the activity and selectivity of electrocatalysts, therefore, the knowledge of the structure‐activity relationship can facilitate the design of efficient catalyst structures for specific reactivity requirements. However, understanding the relationship between a more realistic active surface and its performance is challenging due to the complicated interface microenvironment in electrocatalysis. Herein, we proposed a standard research paradigm to effectively decipher the structure‐activity relationship in electrocatalysis, which is exemplified in the CO₂ electroreduction over SnO₂. The proposed practice has aided in discovering authentic/resting surface states (Sn layer) of SnO₂ accountable for the electrochemical CO₂ reduction reaction (CO₂RR) performance under electrocatalytic conditions, which then is corroborated in the subsequent CO₂RR experiments over SnO₂ with different morphologies (nanorods, nanoparticles, and nanosheets) in combination with in situ characterizations. This proposed methodology is further extended to the SnO electrocatalysts, providing helpful insights into catalytic structures. It is believed that our proposed standard research paradigm is also applicable to other electrocatalytic systems, in the meantime, decreases the discrepancy between theory and experiments, and accelerates the design of catalyst structures that achieve sustainable performance for energy conversion.

Roller bearings are some of the most critical and widely used components in rotating machinery. Appearance defect inspection plays a key role in bearing quality control. However, in real industries, bearing defects are usually extremely subtle and have a low probability of occurrence. This leads to distribution discrepancies between the number of positive and negative samples, which makes intelligent data-driven inspection methods difficult to develop and deploy. This paper presents a small data-driven convolution neural network (SDD-CNN) for roller subtle defect inspection via an ensemble method for small data preprocessing. First, label dilation (LD) is applied to solve the problem of an imbalance in class distribution. Second, a semi-supervised data augmentation (SSDA) method is proposed to extend the dataset in a more efficient and controlled way. In this method, a coarse CNN model is trained to generate ground truth class activation and guide the random cropping of images. Third, four variants of the CNN model, namely, SqueezeNet v1.1, Inception v3, VGG-16, and ResNet-18, are introduced and employed to inspect and classify the surface defects of rollers. Finally, a rich set of experiments and assessments is conducted, indicating that these SDD-CNN models, particularly the SDD-Inception v3 model, perform exceedingly well in the roller defect classification task with a top-1 accuracy reaching 99.56%. In addition, the convergence time and classification accuracy for an SDD-CNN model achieve significant improvement compared to that for the original CNN. Overall, using an SDD-CNN architecture, this paper provides a clear path toward a higher precision and efficiency for roller defect inspection in smart manufacturing.

Electrocatalysis is key to realizing a sustainable future for our society. However, the complex interface between electrocatalysts and electrolytes presents an ongoing challenge in electrocatalysis, hindering the accurate identification of effective/authentic structure–activity relationships and determination of favourable reaction mechanisms. Surface coverage and reconstruction analyses of electrocatalysts are important to address each conjecture and/or conflicting viewpoint on surface-active phases and their corresponding electrocatalytic origin, i.e., so-called structure–activity relationships. In this review, we emphasize the importance of surface states in electrocatalysis experimentally and theoretically, providing guidelines for research practices in discovering promising electrocatalysts. Then, we summarize some recent progress of how surface states determine the adsorption strengths and reaction mechanisms of occurring electrocatalytic reactions, exemplified in the electrochemical oxygen evolution reaction, oxygen reduction reaction, nitrogen reduction reaction, CO₂ reduction reaction, CO₂ and N₂ co-reductions, and hydrogen evolution reaction. Finally, the review proposes deep insights into the in situ study of surface states, their efficient building and the application of surface Pourbaix diagrams. This review will accelerate the development of electrocatalysts and electrocatalysis theory by arousing broad consensus on the significance of surface states.

Electrochemical reactions are pivotal for energy conversion and storage to achieve a carbon-neutral and sustainable society, and optimal electrocatalysts are essential for their industrial applications. Theoretical modeling methodologies, such as density functional theory (DFT) and molecular dynamics (MD), efficiently assess electrochemical reaction mechanisms and electrocatalyst performance at atomic and molecular levels. However, its intrinsic algorithm limitations and high computational costs for large-scale systems generate gaps between experimental observations and calculation simulation, restricting the accuracy and efficiency of electrocatalyst design. Combining machine learning (ML) is a promising strategy to accelerate the development of electrocatalysts. The ML-DFT frameworks establish accurate property–structure–performance relations to predict and verify novel electrocatalysts' properties and performance, providing a deep understanding of reaction mechanisms. The ML-based methods also accelerate the solution of MD and DFT. Moreover, integrating ML and experiment characterization techniques represents a cutting-edge approach to providing insights into the structural, electronic, and chemical changes under working conditions. This review will summarize the DFT development and the current ML application status for electrocatalyst design in various electrochemical energy conversions. The underlying physical fundaments, application advancements, and challenges will be summarized. Finally, future research directions and prospects will be proposed to guide novel electrocatalyst design for the sustainable energy revolution.