251results about How to "Meet speed requirements" patented technology

The invention discloses a method for accelerating a GPU directed at a deep learning super-resolution technology. The method conducts concurrent processing on all the steps of a super-resolution technology which is based on deep learning and a convolutional neural network, and operates on a GPU. The concurrent processing refers to conducting concurrent task dividing on convolutions of the super-resolution technology which is based on deep learning and the convolutional neural network into millions of micro-tasks which are irrelevant to one another and can be concurrently executed in any order so as to fully exhibit the super-strong calculating capability of the GPU. Further, the method uses features of a GPU memory to cache convolution nuclear data and input image data to a shared memory and a register so as to substantially optimize calculating speeds of the convolutions. The method integrates the convolutions and non-linear layers. The method selects the optimal method for the sizes of different convolutions. According to the invention, the method accelerates the high quality super-resolution method to meet velocity requirements for processing videos, and does not cause any image quality loss.

The invention provides a method for plastically manufacturing a high-purity magnesium alloy absorbable stent, relating to a metal stent manufacturing method for the field of intravascular interventional treatment. The method solves the problem of angiosclerosis caused by low yield strength, poor plasticity, poor corrosion resistance and easy damaged cells existing in the prior absorbable stent. The method comprises the following steps: preparing a high-purity Mg-Al-Zn or Mg-Zn-Ca alloy ingot; refining crystalline grains through heat extruding, warm extruding and cold extruding; machining an alloy blank which is extruded for three times into a tube blank; processing the tube blank by multi-pass cold drawing process on the tube blank; machining the thin walled tube into a stent; putting thestent into fluid polishing solution for fluid polishing and burr removing. The method greatly improves the corrosion resistance, yield strength and plasticity, meets the clinical requirement on resorbing speed and time of the intravascular stent, and eliminates the problem of angiosclerosis; and the compositions of the high-purity magnesium alloy stent have higher safety and reliability.

The invention relates to the industrial automation control technology field and relates to a mixed redundancy programmable control system based on equipment redundancy and network redundancy. The mixed redundancy programmable control system comprises a main controller mixed redundancy system and a secondary controller mixed redundancy system, and four levels of redundancy, such as double-Ethernet-equipment redundancy, Ethernet equipment redundancy, double-serial-port serial network redundancy and serial equipment redundancy are included. A main controller is a controller of a double-Ethernet port, four-line mechanism realizes mixed redundancy of the main controller, two switching modes of a normal mode and a rapid mode are provided through matching of an Ethernet mode and an I/O mode, and resource synchronization realizes smooothness of role switching. A secondary controller is a double-serial-port controller, according to a mapping relation between main and secondary equipment, one block of redundancy equipment is added to the secondary equipment, so dual redundancy of a serial communication network and equipment is realized. The mixed redundancy programmable control system is advantaged in that cost is low, four-line mechanism and four levels of redundancy are employed to realize mixed redundancy, reliability is high, and different switching modes satisfy different production technology requirements.

The invention discloses a zero copy network message sending method, which belongs to zero copy method fields that the memory of a user space is directly mapped into a DMA buffer of the network card. The invention comprises the operation steps as follows: A. the device interface is opened; B. a message buffer in the memory of the user space is applied; C. a buffer control structure is constructed; D. the information from the buffer control structure is received through mapping of the memory; E. the sent message data is directly DMA transferred into the message sending buffer; and F. the application program processes the message data. The invention is applicable for sending messages without zero copy protocol stack in high-speed network.

This invention provides a user dispatch method and a system in an uplink strengthened link including: a base station tests the rate of GBR and non-GBR users to be dispatched and their requirements on rate, when the rate of the GBR user does not reach that required by the GBR user, then the base station increases the rate of the GBR user in priority. This invention releases part load from the non-GBR user to increase load of the GBR user.

The invention discloses a hydraulic system of a hydraulic damper test bench, comprising the hydraulic damper test bench and a hydraulic control system. The hydraulic control system comprises a large-level test oil cylinder and a small-level test oil cylinder which are arranged on the hydraulic damper test bench, a central frame is arranged on the hydraulic damper test bench between the large-level test oil cylinder and the small-level test oil cylinder, and a large-level hydraulic damper and a small-level hydraulic damper can be respectively clamped between the central frame and the large-level test oil cylinder as well as between the central frame and the small-level test oil cylinder. In the invention, the test requirements of the large-level hydraulic damper product and the small-level hydraulic damper product are satisfied respectively with the utilization of the large test oil cylinder and the small test oil cylinder, thereby the test accuracy of the product is enhanced; the tensile and compress test requirements and the speed requirements of the hydraulic dampers are satisfied with the utilization of a proportional reversing valve; the test loading requirements in different specifications of the hydraulic dampers are satisfied with the utilization of a proportional relieve valve; and with the utilization of a locking oil cylinder and a pressure keeping loop, reliable locking of the center frame of the test bench is ensured, and continuous working of the hydraulic damper test bench can be also ensured.

A compressor starting torque converter method and apparatus for high power rotating equipment strings includes a compressor starting torque converter (CSTC) (16) and gearing to make the input and output speed conform to the speed and power requirements of the at least one compressor (4) at the end of the string. The string also includes a prime mover (10), either a motor or gas turbine with a starter motor. The CSTC is driven by a prime mover that has been geared down (via 18) to an appropriate speed for efficient power transfer, followed by a gear increasing unit (20) to allow the output of the CSTC to be increased to conform to the necessary requirements of a high speed compressor. The gearing can be two separate units with their own housings, or incorporated in a single housing with the CSTC. The CSTC may be a CSTC used in pressurized starts high compressor load strings for LNG refrigeration service.

The invention relates to a rapid calibration and correction method and a TOF camera thereof, and the method comprises the following steps of (1), collecting the related data of the TOF camera at a fixed frequency, including a depth value image, an integral time value, and real distance value data; (2) processing the data obtained in the step (1), and generating related calibration parameters including a global deviation value, a weight lookup table and an FPPN lookup table; and (3) according to the related calibration parameters generated in the step (2), performing distance correction on theTOF camera, including preprocessing, lighting correction and FPPN correction. The method is simple in principle and simple and convenient to operate, can ensure that the TOF camera has higher measurement precision, can meet the speed requirement in the correction process, and is very suitable for engineering application.

The invention discloses an apparatus and method for testing contact resistance of a conductive slip ring automatically. A low-resistance tester is arranged outside the apparatus. The apparatus comprises a conductive slip ring, a plurality of relays, a relay drive circuit, an automatic control device, and a plurality of leads. The number N of the relays is twice the number M of channels of the conductive slip ring; the N relays are connected with all M channels of the conductive slip ring to form switching of 2N testing channels; each four adjacent relays complete switching of two four-wire measurement channels. The output terminal of the relay drive circuit is connected with the relays and the input terminal is connected with the automatic control device. According to the invention, with the simple relay switching circuit, rapid testing of contact resistances of multiple paths can be realized; and the individual test error can be eliminated.

The invention discloses an LED display screen fault self-recovery control system. According to the control system, a storage unit is divided into two independent storage areas, one storage area is used for storing a program currently running on an LED display screen and display data of the LED display screen, the other storage area is used for storing a backup program, when the program currently running on the LED display screen has a fault or is updated, and the LED display screen automatically switches to the other storage area so that the backup program can run and the display data can be stored. The control system carries out automatic switching on the programs, a receiving card can be replaced without powering off the whole LED display screen, normal work of the LED display screen is not influenced, and the problems that in the prior art, an LED display screen control system cannot be updated on line when the programs are updated, and cannot recover by itself when having a grogram fault or data fault are solved.

The invention relates to large-scale semi-solid die forging hydraulic machine, which comprises frame composed of upper beam, lower beam fixed with liftout attachment, slide block and brace linked by tie bar, plant, power system connected with main and side cylinders both fixed on lower beam, and electric system to control power system; wherein, the piston is arranged in cylinder with hollow plunger and its sleeve; the liftout pad is fixed on piston top; a rotation-proof flange is fixed on piston with guide pole connected with travel digital display, a cylinder outlet flange is fixed on cylinder; the guide sleeve for cylinder outlet is in cylinder; both guidance tape and seal ring are embedded in piston wall; and a seal ring leveler and block drop-proof device of X-shaped are arranged on the machine. This invention has advantages composed of high speed, low power, high efficiency and precision.

The invention relates to an energy-saving power generation dispatching optimization method based on direct current power flow. The method comprises the following steps: arranging the start-stop modes and output levels of non-coal burning units and all coal burning units according to the energy-saving dispatching requirement, and forming a feasible initial population; carrying out iterative optimization of unit combinations; selecting a plurality of better solutions, and randomly generating a plurality of feasible solutions; solving a power balance equation by a direct current power flow method, and carrying out iterative optimization of the direct current power flow; selecting a plurality of better solutions, and randomly generating a plurality of feasible solutions; solving the power balance equation by a newton-raphson power flow method, carrying out iterative optimization of the power flow, and optimizing the start-stop modes and output levels of the units; and finally selecting the optimum solution. In the invention, a genetic algorithm is adopted for carrying out optimization, and cyclic iteration is carried out by sequentially carrying out optimization of the unit combinations, optimization of the direct current power flow and optimization of the newton-raphson power flow, thus the optimum solution of energy-saving dispatching is obtained, and the requirements on calculation accuracy and speed of a large electric network are met.

The invention publishes a vertical shaft supporting system of the vertical spindle type crusher and its vertical shaft fulcrum arrangement, including the vertical shaft support abutment box body, the vertical shaft installs on the vertical shaft support abutment box body, in the upper extreme and the lower extremity of the vertical shaft support abutment box body separately connects the superior cover and the interior cover; liquid hydrostatic bearing which supports the vertical shaft all install on the support abutment box body and the bearing seat of the superior cover. The invention replaces the rolling bearing of the supporting vertical shaft in the existing vertical spindle type crusher by using the liquid hydrostatic bearing, and uses the characteristic of the liquid hydrostatic bearing to improve the supporting condition of the vertical shaft, thus enhances the rotor speed, the bearing life and to compatibility of the continuance random vibration of the vertical spindle type breaker.

The invention discloses a novel memory-based embedded file system and a realization method thereof. A serial peripheral interface Nand flash (SPI Nand flash) chip is arranged in the system. The method comprises the steps of determining that the system supports the SPI Nand flash chip after the system is initialized; setting a corresponding memory technology device (MTD) layer interface during file processing, wherein the MTD layer interface is called when the SPI Nand flash chip is accessed; identifying bad blocks of the SPI Nand flash chip and performing marking and storage, wherein the stored bad block marks are used for troubleshooting the marked bad blocks when the file system accesses the SPI Nand flash chip; performing partitioning processing on the SPI Nand flash chip according to a preset condition; and in a file processing process, accessing the SPI Nand flash chip in a specified partition, thereby finishing file access operation.

A binary image connected domain labeling method for industrial product surface defect detection is provided, comprising: firstly, acquiring a gray level image of a product surface and performing denoising processing, and converting the gray level image into a binary image; then, performing first scanning on the binary image in a manner of scanning once for every three rows from left to right and from up to down, labeling three pixels each time, at the same time of labeling, combining equivalent connected domains and recording representation labels of the connected domains at any time; and finally, performing second scanning on the labeled image, and at the time of performing the second image scanning, duplicating each pixel in a corresponding connected domain by using a representation label of the corresponding connected domain. In a whole process, the number of repeatedly checking pixels is minimized, so that efficiency of a binary image connected domain labeling algorithm is improved; efficiency of detecting a product surface defect is improved, and a requirement of industrial product surface defect detection for a speed is satisfied. Therefore, the method has characteristics of being high in efficiency and strong in practicability.

The invention relates to a video weighted prediction system and a coding and decoding method thereof based on the H.264 standard and the AVS standard in the technical field of image processing, comprising an arithmetic module, a control module, a parameter storage module and a lookup table, wherein, the arithmetic module is connected with the control module so as to transmit a weighted prediction parameter, the control module is respectively connected with the arithmetic module, the parameter storage module and the lookup table so as to transmit the weighted prediction parameter and a control instruction, and the arithmetic module receives a weighted prediction request and outputs a weighted prediction result. The invention uses a set of weighted prediction arithmetic circuit, and input data for forward weighted prediction and rearward weighted prediction does not need to be generated simultaneously, thereby meeting the requirement on speed, saving hardware resources and efficiently accomplishing the H.264 standard and the AVS standard weighted prediction.

The invention discloses a power distribution method in a downlink NOMA system, which is suitable for the downlink NOMA system comprising one base station and MK users, and the base station and the users are provided with single antennas. The base station clusters the users and allocates orthogonal sub-bands to the user clusters; the base station sends the channel condition and the rate requirementof each user; calculates the minimum total power required by each cluster and the minimum total power required by the system; the minimum total power required by the system is used as a constraint condition; constructs a power distribution optimization problem of maximizing the sum of rates of all users in all clusters; The problem is converted into a power distribution optimization sub-problem for maximizing the total rate in a single cluster, each sub-problem is solved, power distribution between clusters is solved on the basis of the sub-problems, and power is distributed to each user in the single cluster with the goal of maximizing the total rate of the cluster according to the inter-cluster power distribution result.

The invention discloses a mechanical boxing training device and a training method for boxing exercises. The training device comprises a device base and a device floating base, wherein a group of locating bolts are inserted on the device floating base; a group of threaded sleeves are arranged on the device floating base; the threaded sleeves are screwed with screwing blocks; floating base springs are arranged at the lower parts of the screwing blocks; a device upper base is arranged at the upper part of the device floating base; a group of vertical pillars are arranged on the device upper base; multi-stage cylinders are arranged on the vertical pillars; the multi-stage cylinders are matched with sliding blocks; telescopic cylinders are arranged at the front parts of the sliding blocks; and the telescopic cylinders are connected to boxing boards by virtue of boxing springs. The mechanical boxing device for the boxing exercises can be firmly placed on the ground, and by regulating the working frequency of the multi-stage cylinders and the telescopic cylinders, boxing practicers can participate in interactive training with the dynamic bowing boards, so that the response capacity of the practicers is trained, with a training effect much better than that of punching a punching bag; and meanwhile, the problem that the practicers are mutually hurt in pair exercises.

The invention discloses a gasoline engine rotation speed control system for a small unmanned helicopter. The control system comprises a rotation speed measuring module, a core controller module, a control signal executing module and a power source management module. The rotation speed measuring module comprises a rotation speed signal acquisition module, a rotation speed calculation module and a data processing module. The invention further discloses a control method of the gasoline engine rotation speed control system for the small unmanned helicopter. The control method includes the following steps that firstly, a high-pressure pulse signal is generated once an engine rotates by a circle, and the high-pressure pulse signal is connected into the rotation speed calculation module through an external capturing interface of a timer in the rotation speed signal acquisition module; secondly, the rotation speed value of the engine is calculated through a T method; thirdly, fused rotation speed data and data of a GPS module are analyzed respectively; fourthly, the difference value between a set rotation speed value and a measured rotation speed value is calculated through a rotation speed control algorithm; and fifthly, the rotation speed of the engine is changed. The gasoline engine rotation speed control system for the small unmanned helicopter and the control method of the gasoline engine rotation speed control system for the small unmanned helicopter have the beneficial effects of being good in extensibility, wide compatibility and the like.

The invention discloses a single stream and double stream switching method based on an effective speed for use in beam forming, relates to a single stream and double stream switching method, and aims to solve the problems that a data transmission speed in single-stream or double-stream beam forming is low, the fairness among users cannot be ensured, and transmission speeds of specific users cannot reach a target speed. The method is implemented by the following steps: I, acquiring a channel matrix defined in the description of all receiving antennas when a user k is on RBn through feedback information of the user k; II, selecting a number defined in the description of the receiving antennas actually used by the user k on an nth resource block, and determining the actually-used receiving antennas according to the number defined in the description; III, establishing a corresponding relation rho<k, n> between the user k of the actually-used receiving antennas and the RBn; and IV, calculating an effective speed defined in the description by the corresponding relation rho<k, n> between the user k of the actually-used receiving antennas and the RBn. The method is applied to the field of single stream and double stream switching.

The invention discloses a flexible connection component, comprising a connection disc, a flange short shaft, a plurality of rubber joint components, wherein the flange short shaft is arranged at an end surface side of the connection disc, the plurality of rubber joint components are uniformly distributed on the end surface of the connection disc, and one end of each rubber joint component is fixed to the connection disc, the other end is fixed with the flange short shaft. The invention has simple structure and excellent displacement compensation performance, and has the advantages of free of lubrication and maintenance, low noise and the like. The invention is applicable to the fields of mechanics, vehicles, shipping and the like.