Search

To enhance the accuracy of theft detection for electricity consumers, this paper introduces a novel strategy based on the fusion of the dual-time feature and deep learning methods. Initially, considering electricity-consumption features at dual temporal scales, the paper employs temporal convolutional networks (TCN) with a long short-term memory (LSTM) multi-level feature extraction module (LSTM-TCN) and deep convolutional neural network (DCNN) to parallelly extract features at these scales. Subsequently, the extracted features are coupled and input into a fully connected (FC) layer for classification, enabling the precise detection of theft users. To validate the method’s effectiveness, real electricity-consumption data from the State Grid Corporation of China (SGCC) is used for testing. The experimental results demonstrate that the proposed method achieves a remarkable detection accuracy of up to 94.7% during testing, showcasing excellent performance across various evaluation metrics. Specifically, it attained values of 0.932, 0.964, 0.948, and 0.986 for precision, recall, F1 score, and AUC, respectively. Additionally, the paper conducts a comparative analysis with mainstream theft identification approaches. In the comparison of training processes, the proposed method exhibits significant advantages in terms of identification accuracy and fitting degree. Moreover, with adjustments to the training set proportions, the proposed method shows minimal impact, indicating robustness.

Microbes play a key role in reef dynamics, mediating the competition between scleractinian corals and benthic algae; however, major shifts in bacterial communities among coral species in response to increases in the abundance of algae are not well understood. We investigated the taxonomic composition of coral-associated microbial communities under algae-overgrowth conditions using 16S rRNA gene sequencing. The results showed that non-algal (i.e., healthy) tissue (HH) had lower bacterial abundance and diversity than tissue collected from the coral–algae interface boundary (HA) and areas of algae growth (AA). Specifically, the HA and AA samples had higher relative abundances of Saprospiraceae, Rhodobacteraceae, and Alteromonadaceae. Compared with Platygyra sp. and Montipora sp., the physiological response of Pocillopora sp. was more intense under algae-induced stress based on microbial gene function prediction. Our results indicate that algal pressure can significantly alter the microbial community structure and function of coral ecosystems. Our data thus provide new insight into the relationship between corals and their microbiome under environmental stress.

Protease-producing bacteria play a vital role in degrading organic nitrogen in marine environments. However, the diversity of the bacteria and extracellular proteases has seldom been addressed, especially in communities of coral reefs. In this study, 136 extracellular protease-producing bacterial strains were isolated from seven genera of scleractinian corals from Luhuitou fringing reef, and their protease types were characterized. The massive coral had more cultivable protease-producing bacteria than branching or foliose corals. The abundance of cultivable protease-producing bacteria reached 106 CFU g−1 of coral. Phylogenetic analysis of 16S rRNA gene sequences revealed that the isolates were assigned to 24 genera, from which 20 corresponded to the phyla Firmicutes and Proteobacteria. Bacillus and Fictibacillus were retrieved from all coral samples. Moreover, Vibrio and Pseudovibrio were most prevalent in massive or foliose coral Platygyra and Montipora. In contrast, 11 genera were each identified in only one isolate. Nearly all the extracellular proteases from the bacteria were serine proteases or metalloproteases; 45.83% of isolates also released cysteine or aspartic proteases. These proteases had different hydrolytic ability against different substrates. This study represents a novel insight on the diversity of cultivable protease-producing bacteria and their extracellular proteases in scleractinian corals.

β-glucosidases can produce gentiooligosaccharides that are lucrative and promising for the prebiotic and alternative food industries. However, the commercial production of gentiooligosaccharides using β-glucosidase is challenging, as this process is limited by the need for high thermal energy and increasing demand for the enzyme. Here, a putative β-glucosidase gene, selected from the coral microbial metagenome, was expressed in Escherichia coli. Reverse hydrolysis of glucose by Blg163 at pH 7.0 and 40 °C achieved a gentiooligosaccharide yield of 43.02 ± 3.20 g·L−1 at a conversion rate of 5.38 ± 0.40%. Transglycosylation of mixed substrates, glucose and cellobiose, by Blg163 consumed 21.6 U/0.5 g glucose/g cellobiose, achieving a gentiooligosaccharide yield of 70.34 ± 2.20 g·L−1 at a conversion rate of 15.63%, which is close to the highest yield reported in previous findings. Blg163-mediated synthesis of gentiooligosaccharides is the mildest reaction and the lowest β-glucosidase consumption reported to date.