Search

Sonic boom suppression is a key technology that must be broken through in the development of a new generation of supersonic civil aircraft. The reasonable design of configuration parameters can make the aircraft have good sonic boom characteristics. In order to break through the bottleneck of evolutionary algorithm in optimizing large-scale design variables, a hierarchical optimization method based on data mining is proposed, and the decision tree (DT) algorithm based data mining is used to extract the design knowledge, obtain the hierarchical information of design variables, and guide the configuration optimization of low sonic boom aircraft. For a low boom supersonic aircraft, five configuration parameters, including sweep angle, aspect ratio, taper ratio, dihedral angle and fuselage slenderness ratio, are selected as design variables to carry out the numerical experiments of hierarchical optimization, and the method is performed with comparison verification with the integrated optimization method. The results show that the hierarchical optimization method can obtain the optimal solution consistent with the integrated optimization method, and the convergence speed of hierarchical optimization method is significantly faster than that of integrated optimization method, and the performance of different optimization processes is more robust.

The airfoil is the foundation of an aircraft, and its characteristics have a significant impact on those of the aircraft. Conventional airfoil design mainly focuses on improving aerodynamic performance, while flying wing airfoil designs should also consider layout stability and stealth performance. The design requirements for an airfoil vary with its position on the flying wing layout aircraft based on corresponding spanwise flow field characteristics. By analyzing the spanwise flow characteristics of the flying wing, partition design models for flying wing airfoils were established in this study, and a series of flying wing airfoil designs that consider aerodynamics and aerodynamic/stealth were implemented. Then, the designed airfoils were configured on a three-dimensional X-47B layout for testing and verification. The results showed that the aerodynamic design and the aerodynamic/stealth design exhibited significant improvements in terms for aerodynamic and longitudinal trimming characteristics. However, the cruise drag performance of the aerodynamic/stealth design was slightly worse than that of the aerodynamic design, although the longitudinal moment trimming characteristics were basically the same. The stealth characteristics of the aerodynamic/stealth design had significant advantages, indicating that there were weak contradictions between the aerodynamic, stealth, and trimming requirements in the design of the flying wing. [ABSTRACT FROM AUTHOR]

Nanopore sequencers can enrich or deplete the targeted DNA molecules in a library by reversing the voltage across individual nanopores. However, it requires substantial computational resources to achieve rapid operations in parallel at read-time sequencing. We present a deep learning framework, NanoDeep, to overcome these limitations by incorporating convolutional neural network and squeeze and excitation. We first showed that the raw squiggle derived from native DNA sequences determines the origin of microbial and human genomes. Then, we demonstrated that NanoDeep successfully classified bacterial reads from the pooled library with human sequence and showed enrichment for bacterial sequence compared with routine nanopore sequencing setting. Further, we showed that NanoDeep improves the sequencing efficiency and preserves the fidelity of bacterial genomes in the mock sample. In addition, NanoDeep performs well in the enrichment of metagenome sequences of gut samples, showing its potential applications in the enrichment of unknown microbiota. Our toolkit is available at https://github.com/lysovosyl/NanoDeep. [ABSTRACT FROM AUTHOR]