Search

There is a complex transition phenomenon in the flow field of a typical hypersonic vehicle, which has a significant impact on the performance of the vehicle. The effects of Mach number and wall temperature on the transition of HyTRV were studied by numerical simulation methods. The self-developed software of the research group was used to carry out numerical calculations. The range of Mach number was 3~8, and the range of wall temperature was 150~900 K. Firstly, the hypersonic corrections of the γ-$\mathop R\limits^ \sim $eθt transition model and the SST turbulence model were carried out. The pressure gradient coefficient correction and the high-speed cross-flow correction were introduced into the γ-$\mathop R\limits^ \sim $eθt transition model, and the compressibility corrections of the closure coefficients β* and β of the SST turbulence model were carried out. Then, the grid independence verification was carried out, and the modified numerical method and software platform were confirmed by comparing with experimental results. Finally, the effects of Mach number and wall temperature on the transition law of the HyTRV boundary layer were studied. The results show that the transition area is mainly concentrated on both sides of the upper surface and the center line of the lower surface. With the increase of the incoming Mach number, the starting position of transition on the upper and lower surfaces is greatly backward, and the turbulent zone is greatly reduced, but it still exists. At the same time, the friction coefficient of the laminar flow zone on the upper surface increases continuously, and the friction coefficient of the turbulent zone on the lower surface decreases. As the wall temperature increases, the starting position of transition on the upper and lower surfaces shifts forward, then rapidly shifts backward, and finally the turbulent zone almost disappears.

In order to reduce the processing volume of multi-camera real-time video data in smart city scenarios, a video collaborative analysis method based on machine learning algorithms at the edge was proposed.Firstly, for the important objects detected by each camera, different key windows were designed to filter the region of interest (RoI) in the video, reduce the video data volume and extract its features.Then, based on the extracted data features, the same objects in the videos from different cameras were annotated, and a strategy for calculating the association degree value between cameras was designed for further reducing the video data volume.Finally, the GC-ReID algorithm based on graph convolutional network (GCN) and re-identification (ReID) was proposed, aiming at achieving the collaborative analysis of multi-camera videos.The experimental results show that proposed method can effectively reduce the system latency and improve the video compression rate while ensuring the high accuracy, compared with the existing video analysis methods.

To ensure the safety of life and property of residents on landslides and reveal the deformation and failure law of masonry buildings on landslides, it is necessary to carry out a simulation study on the deformation and failure process of masonry buildings under the action of landslide deformation. A refined model of a masonry building was established based on the contact separation modeling method in ABAQUS, and the viscous contact interface was used to simulate the effect of mortar in the wall. Comparing the results of the numerical simulation with the physical model test results, the main conclusion is as follows: through the comparison of macroscopic deformation characteristics and microscopic stress and strain data obtained from physical and numerical model tests, the load-strain curve of the numerical model test agrees well with that of the physical model test, and their strain clouds are consistent, proving the validity of the numerical simulation method. The law of crack propagation and strain distribution of building walls in landslide tension areas is revealed, which can provide a basis for the protection design of masonry buildings in landslide tension areas.

Multiple kernel learning(MKL) aims to find an optimal consistent kernel function.In the hierarchical multiple kernel clustering(HMKC) algorithm,the sample features are extracted layer by layer from high-dimensional space to maximize the retention of effective information,but the information interaction between layers is ignored.In this model,only the corresponding nodes in the adjacent layer will exchange information,but for other nodes,it is isolated,and if the full connection is adopted,the diversity of the final consistence matrix will be reduced.Therefore,this paper proposes a hierarchical multiple kernel K-Means(SCHMKKM) algorithm based on sparse connectivity,which controls the assignment matrix to achieve the effect of sparse connections through the sparsity rate,thereby locally fusing the features obtained by the distillation of information between layers.Finally,we perform cluster analysis on multiple data sets and compare it with the fully connected hierarchical multiple kernel K-Means(FCHMKKM) algorithm in experiment.Finally,it is proved that more discriminative information fusion is beneficial to learn a better consistent partition matrix,and the fusion strategy of sparse connection is better than the strategy of full connection.

In order to improve the utilization of wind energy and solve the problem of single rotor rotation of existing switched reluctance generators, a new type of deflectable dual stator Switched Reluctance Wind generator is proposed. Firstly, based on the electromagnetic-solid mechanics module, the electromagnetic field, flux density and magnetostrictive density of stator structure are analyzed theoretically by establishing the magneto-solid coupling numerical model. Secondly, the finite element platform is used to model the generator intuitively. based on the theory of mechanics, the basic equation of electromagnetic solid mechanics considering magnetostrictive effect is established. Finally, the finite element method is used to simulate the internal and external stators of the generator and the rotation and deflection of the rotor. The stress distribution and the corresponding vibration displacement are obtained. The simulation results show that: The vibration displacement under magnetostrictive effect belongs to nanometer level and the stress is small, so the influence on generator operation can be neglected. The analysis method of magneto-solid coupling provides theoretical reference in analyzing motor vibration and further optimizing generator parameters.

Aiming at the shortcomings of false alarm and leakage alarm in the existing fire detection system and the complicated wiring,high cost and poor flexibility of the traditional wired fire detection system,a set of home fire detection system based on wireless multi-sensor data fusion was designed,to detect the fire safety hazards caused by kitchen equipment aging,electrical wiring aging or electrical equipment aging.The design of multi-sensor data fusion algorithm,fire detection system hardware framework and software implementation were described in detail.The simulation results show that the scheme has the characteristics of high accuracy and flexibility.

To evaluate the quality differences of four mainstream species of Berberidis Cortex,~1H-NMR metabolomics was applied to detect its primary and secondary metabolites, and the partial least squares discriminant analysis and analysis of variance were integrated to screen differential metabolites between species. Furthermore, diabetic rat model was established by high fat diet and streptozotocin to assess differences in antidiabetic activities among the four species. Sixteen compounds were simultaneously detected and identified, including alkaloids, organic acids, carbohydrates and amino acids. Interspecific difference markers were revealed as magnoflorine, jateorhi-zine, bufotenidine and saccharose for the first time. Berberis vernae and B. kansuensis presented superior activities on reducing blood glucose level, improving insulin resistance, increasing insulin sensibility and anti-inflammation. B. dictyophylla showed moderate antidiabe-tic effect, while B. diaphana rendered inferior antidiabetic capacity. Based on the contents of four differential markers and the results of antidiabetic activity evaluation, the quality of four Berberidis Cortex species was ranked as B.vernae≈B.kansuensis>B. dictyophylla>B. diaphana. These results provided references for species collation, quality standard establishment and exploitation of Berberidis Cortex. The antidiabetic activities of B. vernae and B. kansuensis as well as their mechanisms of action merit further study in the future.