The development of efficient photocatalysts for water purification remains challenged by rapid charge recombination and limited surface reactivity. Here, we report a rationally designed PHI/W18O49 heterojunction photocatalyst, integrating interfacial W-N bonds and oxygen vacancies to synergistically overcome these limitations. The optimized PHI/W18O49 heterojunction exhibits significantly enhanced visible-light absorption, efficient spatial charge separation, and improved surface reaction kinetics. Consequently, it demonstrates superior and versatile photocatalytic performance in degrading organic contaminants, along with stable recyclability and consistent performance in real water matrices, highlighting its potential for practical applications. Mechanistic studies combining DFT calculations and experimental characterization reveal that interfacial W-N bonds direct S-scheme electron transfer, while oxygen vacancies enhance reactant adsorption, jointly boosting charge separation and surface reactivity. This work establishes a design paradigm that couples' defect-enhanced surface adsorption with directed interfacial electron flow, offering an effective strategy for developing robust photocatalysts for practical environmental remediation.
Starch retrogradation, driven by reassociation and recrystallization of gelatinized amylose and amylopectin chains, induces undesirable firming, syneresis, and sensory deterioration, ultimately shortening the shelf-life of starch-based foods. Enzymatic modification has emerged as a promising route to retard retrogradation, but its efficacy critically depends on the structural features targeted by the enzyme, the susceptibility of the starch substrate, and the accessibility of catalytic sites. This review systematically interprets the structure-dependent mechanisms underlying enzymatic retrogradation regulation, covering the differential regulatory patterns of various enzyme systems. This review further clarifies the essential distinction between substrate structural susceptibility and enzyme catalytic accessibility, and summarizes advanced modification strategies based on porous starch construction and controlled starch swelling to optimize enzyme-substrate interfacial contact efficiency. In addition, mainstream detection techniques are compared based on the different levels of the studied starch structure. Finally, the practical applicability of enzymatic anti-retrogradation technologies in diverse staple and processed starch-based food systems are also summarized. We propose that the key design principle is controlled structural remodeling enabled by enzyme action within a defined processing window, rather than indiscriminate starch degradation. This work provides systematic theoretical insights and practical guidance for the quality stabilization and shelf-life extension of starch-based food products.
Guangxi cassava (Manihot esculenta Crantz) was used with medium-temperature Daqu as the fermentation agent to enhance the nutritional value and physicochemical properties of the products through solid-state fermentation. Using an orthogonal experimental design, the optimal fermentation conditions were determined as follows: inoculation volume of 8%, moisture content of 60% (m/m), and fermentation time of 48 h. Under these optimized conditions, the amino acid and reducing sugar content of fermented cassava reached 37.51 and 18.54 mg/g, respectively, and the contents of leucine, valine, and phenylalanine were significantly increased by 300.00%, 231.03%, and 120.21%, respectively. Additionally, the anti-nutritional factors linustatin and linamarin were effectively reduced. Metagenomic sequencing results revealed that Enterococcus, Enterobacter, and Lactococcus were the dominant bacterial genera, with relative abundances of 52.15%, 18.28%, and 14.40%, respectively. Amino acids and their derivatives, along with organic acids, emerged as the main differentially expressed metabolites. Among the volatile metabolites in fermented cassava, palmitic acid, elaidic acid, and linoleic acid were the most abundant, reaching concentrations of 51.91, 48.35, and 29.72 µg/g, respectively. Overall, this study demonstrated that medium-temperature Daqu could serve as an effective starter for solid-state fermentation of cassava, effectively improving its nutritional composition, flavor characteristics, and safety profile.
During Tunnel Boring Machine (TBM) operations, appropriate cutterhead loads regulation is crucial for ensuring both efficiency and safety. However, accurate loads prediction remains challenging due to intrinsic high-frequency fluctuations and potential gradient explosion issues. To address these challenges, this study proposes the TPE-MLSTM-ICA model, which significantly improves prediction accuracy through optimization strategies tailored to high-frequency fluctuation characteristics. The model effectively captures load fluctuation dynamics and parameter coupling relationships through a Mogrifier-enhanced LSTM and an Improved Cross-Attention (ICA) mechanism, while employing a weighted Huber loss function to enhance stability. Optimized via the TPE algorithm, the model achieves MAPE values of 6.03% for torque and 0.74% for thrust in limestone strata within the Central Yunnan Water Diversion Project, requiring a training time of only 1342 s. Cross-strata tests demonstrate that the model maintains excellent predictive performance in both breccia and basalt formations, with thrust prediction MAPE consistently remaining below 1%, thereby confirming its outstanding engineering applicability.
Industrial fault diagnosis often faces the challenge of limited fault samples. Most existing closed-set diagnosis approaches focus on weighting and aligning source domains with the target domain. However, the domain shifts among multiple source domains are often overlooked. To address these issues, this paper proposes a Task-Structured Prototype Evolution with Dual-Stage Reconstruction (TSPE-DSR) framework for multi-source cross-domain few-shot closed-set fault diagnosis. First, a cross-domain task division mechanism is designed to construct meta-tasks from multiple source domains. The mechanism iteratively assigns pseudo-source and pseudo-target domains to form auxiliary tasks. These tasks exploit shared knowledge across domains and facilitate rapid adaptation to the target domain. Second, a dual-stage attention prototype reconstruction network is developed to model prototype evolution. A self-attention mechanism is introduced to achieve a progressive transition from embedding learning to prototype reconstruction. This process enables adaptive transfer of source-domain prototypes and effectively reduces cross-domain discrepancies. Finally, a joint training strategy is adopted by combining weighted prototype similarity loss and classification loss. This strategy improves classification performance, alleviates overfitting, and enhances model robustness. The effectiveness of TSPE-DSR is validated on the Tennessee Eastman Process(TEP), Case Western Reserve University(CWRU), and Aluminum Electrolysis Process(AEP) datasets. Experimental results demonstrate that TSPE-DSR outperforms existing methods in classification performance, achieving average accuracy improvements of 5.52%, 3.12%, and 4.79% over the strongest baseline on the TEP, CWRU, and AEP datasets, respectively. Our code is at https://github.com/Luce-hub/DAPRN.
