Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

3results about How to "Shorten the progress" patented technology

Multidisciplinary simulation coupling method, device and computer equipment for aero-engine

ActiveCN120068691BEffective simulation of matching situationsAvoid the problem of large simulation deviationsSustainable transportationDesign optimisation/simulationSimulationAero engine
The application relates to a multidisciplinary simulation coupling method, device and computer equipment of an aero-engine. The method comprises the following steps: acquiring simulation running models of each discipline of the aero-engine and associated mechanism information therebetween, and identifying coupling information of each coupling type between the disciplines; based on the coupling information of each coupling type between the disciplines, constructing each level coupling mechanism model of the aero-engine through a hierarchical coupling modeling strategy, thereby performing running association processing on the simulation running models of each discipline to obtain a simulation coupling model of the aero-engine; collecting simulation working condition information and simulation condition information of the aero-engine, and simulating the running process of the aero-engine through the simulation coupling model to obtain simulation running results of the aero-engine; and based on the simulation running results of the aero-engine, identifying simulation running information of each discipline of the aero-engine. The method can improve the accuracy and efficiency of accounting for actual course learning hours of users.
Owner:TSINGHUA UNIVERSITY

A kind of tea the screening device for purifying tea theaflavins

This invention discloses a sieving device for purifying theaflavins in tea, comprising a high-temperature and high-pressure treatment tank. The top and bottom of the high-temperature and high-pressure treatment tank are equipped with connecting pipes. The sieving device also includes a fixing frame, which is fixedly connected to the surface of the high-temperature and high-pressure treatment tank. A motor is fixedly mounted on the surface of the fixing frame, and the drive shaft of the motor is fixedly connected to a connecting seat. This invention uses a second electric push rod to retract, causing a fixing pin to move horizontally and disengage from the mounting post. An electric push rod third, located on the back of the connecting seat, extends. When the filter tank body rotates to below the high-temperature and high-pressure treatment tank, the extract can be filtered again by the filter assembly. Therefore, the position of the filter assembly can be adjusted according to actual needs to directly filter and sieve the extract, saving time spent during transportation and shortening the overall tea processing time, thus improving the practicality of the device.
Owner:JIANGSU DEHE BIOTECH

Contract recursion method for distributed tensor data updating

The invention discloses a contract recursion method for distributed tensor data updating, and relates to the technical field of computer data processing, and the method comprises the following steps: initializing a dynamic adaptive contract cluster, sensing and aggregating the dynamic state of the cluster in real time, executing a contract recursion update cycle, carrying out fault tolerance and state restoration in a recursion process, and carrying out iteration termination judgment. According to the invention, through combination of the dynamic adaptive contract and real-time cluster state perception, intelligent recursion of distributed tensor update is realized, the efficiency and stability of large-scale training are significantly improved, and a communication strategy can be dynamically optimized according to a network and computing power state; state consistency in a complex environment is ensured, conflicts are accurately resolved through version vectors and intelligent arbitration, and elegant fault tolerance is realized by utilizing prediction compensation; and meanwhile, the method has high adaptability, parameters can be optimized on line to reduce manual tuning cost, and robust and efficient support is provided for distributed machine learning.
Owner:MINASH (SHANGHAI) ROBOT TECHNOLOGY CO LTD +1