Search

The trend of low breastfeeding rates increases the demand for infant milk formula (IMF) worldwide, but the use of IMF may be one of the causes of bacterial infections in infants. Complete sterility in the whole production line of IMF cannot be guaranteed; therefore, it is necessary to closely monitor the microbial content in the process. In the present study, an IMF powder production line based on the wet mixing process was sampled at 27 suspicious points in spring and summer to analyze the bacterial diversity by high-throughput sequencing. We found that 70 and 69 different bacterial phyla were present in spring and summer samples, respectively, with Proteobacteria and Firmicutes being the dominant phyla (>80% relative abundance). Moreover, 13 dominant genera each were present in spring (e. g., Pseudomonas and Lactococcus) and summer (e. g., Pseudomonas, Bacillus, and Streptococcus). Samples associated with workers showed higher bacterial species diversity (Shannon index) and richness (Chao1 index) in summer than in spring. The bacterial community composition showed high similarity between liquid milk after pasteurization and concentrated milk after evaporation. The potential bacterial pathogens were identified as Pseudomonas aeruginosa in spring and Acinetobacter baumannii in summer. Through retrospective analysis of the two opportunistic pathogens identified, it was found that the workshop environment was the potential contamination point in spring, whereas the auxiliary ingredients were the potential source of contamination in summer. The results highlight the effect of season on bacterial diversity associated with the production process of IMF and are useful in controlling the microbial quality and safety of infant dairy products.

With the development of technologies such as IoT and 5G, the exponential explosion in the amount of new data has put more stringent requirements on ultrareliable and low-delay communication of services. To better meet these requirements, a resource allocation strategy using deep reinforcement learning in a cloud-edge collaborative computing environment is proposed. First, a collaborative mobile edge computing (MEC) system model, which combines the core cloud center with MEC to improve the network interaction ability, is constructed. The communication model and computation model of the system are considered at the same time. Then, the goal of minimizing system delay is modeled as a Markov decision process, and it is solved by using the deep Q network (DQN) which is improved by hindsight experience replay (HER), so as to realize the resource allocation with the minimum system delay. Finally, the proposed method is analyzed based on the simulation platform. The results show that when the number of user terminals is 80, the maximum user delay is 1150 ms, which is better than other comparison strategies and can effectively reduce the system delay in complex environment.

To address the problem of poor detection performance of existing intrusion detection methods in the environment of high-dimensional massive data with uneven class distribution, a deep learning-based anomaly traffic detection method in cloud computing environment is proposed. First, the fuzzy C -means (FCM) algorithm is introduced and is combined with the general regression neural network (GRNN) to cluster the samples to be classified in the original space by FCM. Then, the GRNN model is trained and the center point is updated using the sample closest to the FCM clustering center until a stable cluster center is obtained. The parameters in FCM-GRNN are optimized using the global optimization feature of the modified fruit fly optimization algorithm (MFOA), and the optimal spread value is found using the three-dimensional search method through an iterative search. Finally, experiments are conducted based on the KDD CUP99 dataset, and the results demonstrate that the detection rate (DR) and false alarm rate (FAR) of the proposed FCM-MFOA-GRNN method are 91% and 1.176%, respectively, which are better than those of the comparison methods. Therefore, the proposed method has good anomaly traffic detection ability.

Aiming at the lack of feature extraction ability of rumor detection methods based on the deep learning model, this study proposes a rumor detection method based on deep learning in social network big data environment. Firstly, the scheme of combining API interface and third-party crawler program is adopted to obtain Weibo rumor information from the Weibo “false Weibo information” public page, so as to obtain the Weibo dataset containing rumor information and nonrumor information. Secondly, the distributed word vector is used to encode text words, and the hierarchical Softmax and negative sampling are used to improve the training efficiency. Finally, a classification and detection model based on the combination of semantic features and statistical features is constructed, the memory function of Multi-BiLSTM is used to explore the dependency between data, and the statistical features are combined with semantic features to expand the feature space in rumor detection and describe the distribution of data in the feature space to a greater extent. Experiments show that when the word vector dimension is 300, compared with the compared literature, the accuracy of the proposed method is improved by 4.232% and 1.478%, respectively, and the F1 value of the proposed method is improved by 5.011% and 1.795%, respectively. The proposed method can better extract data features and has better rumor detection ability.

Type synthesis of mechanical structure is of great significance to the realization of mechanism target function, systematization, and stability of mechanical device. The type synthesis method of multilinkage robot has been given high demands with increasing number of degrees of freedom and high flexibility in special occasions. In order to improve the workspace and flexibility of mechanism, this paper studies the existing type synthesis theory and proposes a type synthesis method of modular combination with virtual rotation centers. Firstly, modular units are built. Secondly, modular units are expanded according to the needed paths. In the end, the expanded modular units are combined to form the kinematic linkages. Based on the proposed method, the configuration design of the aerial working platform and the self-adaptive levelling platform is completed. The stabilities of two platforms are checked by modal analysis. The prototype products are manufactured, respectively, for further verifying validity of the method. The designed aerial working platform with virtual rotation centers can achieve 360° rotating large workspace, more compact mechanical structure, and short arrival time at the same height than the common scissor-type and mast-type aerial working platforms. The designed adaptive levelling platform is tested that ensures the levelling of the upper surface at different inclinations. The method can provide new idea for the mechanism configuration and expand the application scope of new mechanisms.