Beijing University of Technology

www.bjut.edu.cn/
Beijing, China

Beijing University of Technology , also called Beijing Polytechnic University or Bei Gong Da , is recognized as one of the Project 211 universities. The University has established a multidisplinary academic structure, offering a variety of programs and is involved in diversified research in the fields of Science, Engineering, Economics, Management, Liberal Arts, and Law. Wikipedia.

SEARCH FILTERS
Time filter
Source Type

The present invention provides a sector-shaped section of a continuous casting machine for heavy reduction of a solidified terminal of a continuously cast slab and a heavy reduction method. The sector-shaped section of a continuous casting machine comprises an upper frame, a lower frame, an upper driving roller, a lower driving roller, a left driven roller set, a right driven roller set, a screw-down gear and clamping cylinders which are positioned on the sector-shaped section; the clamping cylinders are used for clamping the upper frame and the lower frame and keeping the upper frame and the lower frame at a set interval; the upper driving roller is connected with the screw-down gear; the upper driving roller is connected to the upper frame through a bearing base; the lower driving roller is connected to the lower frame through a bearing base; the left driven roller set and the right driven roller set, are respectively positioned on two sides of the driving rollers; the left driven roller set is used for clamping the cast slab before reduction; the right driven roller set is used for clamping the cast slab after reduction; and the diameter ratio of the driving rollers to the driven rollers is 1.1:1-2:1. The deformation permeability of the present invention is increased, equivalent to the deformation of one rolling pass of a two-roller mill with a relatively large roller diameter, which facilitates improvement on the looseness and segregation of the center area of the continuously cast slab.


Patent
Beijing University of Technology | Date: 2017-04-15

A sandwich structure for enhancement of photoluminescence (PL) from luminescent films and the corresponding preparation method are disclosed. The sandwich structure comprises a support, a luminescent film grown on the support, and a single-layer close-packed microsphere array deposited onto the luminescent film. The microspheres have high transmittance excitation light and emitted light, respectively. The low price of dielectric microspheres is beneficial to industrial applications. The stable chemical properties of dielectric microspheres make PL enhanced in a long term. Both metal and non-metal materials can be used as the support in the sandwich structure. These features significantly improve the technique of PL enhancement for luminescent films.


Patent
Beijing University of Technology and Guizhou Expressway Group Co. | Date: 2017-07-05

The invention relates to an illumination standard calculation method for a tunnel entrance section in daytime based on safe visual recognition. The method includes: (a) setting light environment parameters of the tunnel entrance section; (b) determining a feature position in the tunnel, and measuring the average road surface brightness L of the feature position; (c) placing a target object in the feature position and then making a driver drive at different speeds from outside the tunnel towards the target object to measure visual recognition distances D required by the driver to visually discover the target object under different driving speeds, and measuring the road surface brightness L_(20) at the position where the driver visually discovers the target object and calculating the brightness reduction factor K to be the ratio between L and L_(20); (d) resetting the average road surface brightness L of the tunnel entrance section and repeating steps (b) and (c) to obtain a plurality of different sets of visual recognition distances D and corresponding brightness reduction factors K; (e) using Sigmoid function model to calculate and fit the data of the plurality of sets of D and K to obtain a relational model formula of D and K: _(0) corresponding to a maximum speed limit of the tunnel into the formula to obtain the minimum brightness reduction factor K_(0) of the tunnel entrance section in daytime. The invention also relates to a system for implementing the illumination standard calculation method for a tunnel entrance section in daytime based on safe visual recognition.


Patent
Beijing University of Technology | Date: 2015-12-31

This invention disclosed a visual perception characteristics-combining hierarchical video coding method, comprising: setting priority of visual interest area and setting allocation scheme of video coding resource. The former one is about: due to the richness of video image and selective perception of human visual, visual characteristics saliency of video content has been analyzed from time and space, priority of visual interest area can be labeled. The later one is about improving real-time performance of video coding, while at the same time, quality of video coding and compression efficiency is guaranteed. According to the priority of visual interest area, macroblock coding resource of interest area should be satisfied firstly, to realize hierarchical coding. The video coding scheme of this paper can remit conflict between coding complexity and coding efficiency validly. Compare with H.264/AVC international video coding standard, it can improve 80% of average coding speed, and at the same time it can maintain better quality of video image and lower compression bit-Rate. The video coding performance can be improved dramatically.


The present disclosure provides a full-duplex wireless communication method, an antenna device and a full-duplex wireless communication system. The method includes steps of: within a training period, acquiring a relevant parameter for cancelling a self-interference signal from an antenna itself as a first estimated value, and a relevant parameter for cancelling a cross-interference signal between the antenna and the other antenna as a second estimated value; and within a data transmission period, receiving and transmitting signals simultaneously by the antenna using an identical frequency, cancelling the self-interference signal from the antenna itself in accordance with the first estimated value, and cancelling the cross-interference signal between the antenna and the other antenna in accordance with the second estimated value.


A method for improving resistance of a fusion reactor inner wall to plasma irradiation by using lamination structure comprises: making a material for facing plasma into multiple metal sheets, laminating the multiple metal sheets together according to a direction perpendicular to a wall surface, and combining the laminated multiple metal sheets with a copper substrate. The method relates to the field of nuclear applications, and is applicable to a plasma-facing surface on an inner wall of an apparatus for conducting a fusion reaction by using hydrogen isotopes. Not only does the method reduce aggregation of hydrogen, helium, and isotopes thereof, and the like, under a surface layer of a tungsten-based material, greatly reducing a blistering phenomenon on a surface thereof, but also reduces thermal fatigue cracking damages.


Patent
Beijing University of Technology | Date: 2015-04-30

The present disclosure relates to a method of acquiring neighboring disparity vectors for multi-texture and multi-depth video (e.g., 3D-HEVC video). The method belongs to the area of 3D-HEVC video coding technology. The method includes changing the standard associated with a disparity vector that is first searched as a final disparity vector. By deleting location which is minimum searched in candidate space and time location of the coding unit next to current coding unit to divide candidate space and time location of the coding unit into groups, the method takes searched disparity vector that is combined based on the proportion of adoption rate as final disparity vector. The method improves coding quality and at the same time maintaining origin fast algorithm efficiency. The embodiments of the present disclosure improve coding quality at least 0.05% and at the same time maintain origin fast algorithm efficiency while the decoding time is decreased to 97.1%.


Patent
Beijing University of Technology and Guizhou Expressway Group Co. | Date: 2017-08-09

The present invention relates to a visual efficacy determining method for non-colored objects in different light environments, comprising: a) setting light environment; b) placing an object for performing visual recognition of a driver; c) resetting the light environment, and repeating the step b); d) processing visual recognition information data obtained after experiments, and establishing a correlating relationship between visual recognition time and color temperature, color rendering index and brightness parameters of the light environment; and e) performing visual efficiency analysis according to the visual recognition information data processing results. The present invention further provides a system for implementing the visual efficacy determining method for non-colored objects in different light environments, wherein the system comprises a simulated tunnel middle section subsystem, a testing subsystem and a data processing subsystem. According to the method and the system disclosed by the present invention, visual efficacy analysis is performed on the visual influence of the light environment setting of the illumination of a night road and a tunnel middle section on the driver, and good guidance standards are provided for the reasonable and efficient setting of the color temperature, color rendering index and average brightness of a lighting device.


Patent
Beijing University of Technology and Guizhou Expressway Group Co. | Date: 2017-06-21

This invention relates to an illumination standard calculation method for a tunnel middle section based on safe visual recognition: (a) Setting the light environment of the tunnel middle section; (b) Placing a target object in the tunnel middle section; (c) Making a driver drive a motor vehicle at different speeds toward the target object, and measuring the visual recognition distances D required by the driver to visually discover the target object at different speeds; (d) Resetting the average brightness L of the tunnel middle section and repeating the steps (b) and (c) to obtain a plurality of different sets of visual recognition distances D and corresponding brightness values L; (e) Using the S model to fit the data of the plurality of sets of visual recognition distances D and brightness values L to obtain the formula of the relational model of D and L to be L=0.683/(5.575-In(D)); (f) Substituting a safe stopping sight distance D_(0) corresponding to a maximum speed limit of the tunnel into the model formula to obtain the dynamic minimum brightness value L_(0) required for the tunnel middle section under this tunnel light environment. The method improves the accuracy of safety evaluation of the tunnel middle section brightness, and the method is simple and convenient, and provides a reference basis for the road traffic safety research. The invention also provides a system for implementing this method.


Patent
Beijing University of Technology | Date: 2016-02-08

The present invention relates to a clustering method based on iterations of neural networks, which comprises the following steps: step 1, initializing parameters of an extreme learning machine; step 2, randomly choosing samples of which number is equal to the number of clusters, each sample representing one cluster, forming an initial exemplar set and training the extreme learning machine; step 3, using current extreme learning machine to cluster samples, which generates a clustering result; step 4, choosing multiple samples from each cluster as exemplars for the cluster according to a rule; step 5, retraining the extreme learning machine by using the exemplars for each cluster obtained from step 4; and step 6, going back to step 3 to do iteration, otherwise obtaining and outputting clustering result until clustering result is steady or a maximal limit of the number of iterations is reached. The present invention resolves problems that how to realize clustering of high dimensional and nonlinear data space and that the prior art consumes a larger memory or need longer running time.

Loading Beijing University of Technology collaborators
Loading Beijing University of Technology collaborators