104results about How to "Meet computing needs" patented technology

The invention belongs to the field of electric power system dispatching, and particularly relates to a dynamic probability dispatching method for calculating system operation reliability. A new method based on Benders decomposition is provided for solving the large linear programming problem formed by probability dynamic dispatching. Aiming at the coupling relation among various operation states and according to the idea of decomposition coordination, the original problem is decomposed by the Benders decomposition technology so as to form an iterative solution format formed by a dynamic economic dispatching main problem in a normal operation state and an operation state adjustment sub-problem in an accident operation state; and the solution scale of each optimized calculation is reduced. By the method, the solution speed of problems is increased, and the effective solution of large-scale systems is realized.

The invention relates to the technical field of server architecture, Provided are a device and a method for freely switching GPU server topology. The device includes a CPU board, GPU board, a PCIE board and a switch board. A PCIE Switch is arranged on the switch board. Four PCIE interfaces are arranged on the PCIE Switch, and the interfaces are used to connect with the CPU board, the GPU board andthe PCIE board. The switch board changes connection relationship among the four PCIE interfaces by switching working modes, and realizes changes of a topology structure of the GPU server. By introducing thea switch board and configuring the working mode of the switch board, the device and the method realize dynamic adjustability of a PCIE interconnection structure, so that a user can switch between two topological structures without changing an internal structure of the server under the condition of not opening a case, so as to satisfy computing requirements of different scenes, and improve operation and maintenance efficiency of a data center.

The invention discloses a multi-GPU-based graph data processing system and method. The method comprises the steps of firstly dividing input original graph data through a data division module, and allocating the divided data to GPU devices; secondly allocating different computing tasks to vertexes on the GPU devices according to computing attributes through a task management module, and managing an executive sequence of various tasks; thirdly finishing execution of parallel computing tasks on the GPU devices while ensuring data consistency through an executive module; and finally finishing a data collection process after computing completion through a data collection module. According to the system and the method, a new task allocation and management mechanism is proposed; parallel efficiency is maximized; and communication is masked with computing, so that the problems of high GPU-CPU communication overhead and finite graph data processing scale of an existing GPU-based graph processing system are solved and the efficiency of processing large-scale graph data in a GPU is improved.

The invention discloses a mobile edge computing unmanned aerial vehicle group auxiliary communication method for explosive traffic, which is responsible for providing computing service for users withcomputing requirements on the ground when a network is congested. Firstly, in the prediction part, an improved PSO-BP neural network is used for predicting the calculation demand distribution situation of a target area, and unmanned aerial vehicles supporting mobile edge calculation are deployed as required; secondly, a demand partitioning part is calculated, a target area is partitioned based ona fairness principle, and each unmanned aerial vehicle serves users in a sub-area; and finally, the unmanned aerial vehicle energy optimization part in each sub-region performs joint optimization on the unmanned aerial vehicle calculation frequency and the unmanned aerial vehicle flight path on the premise of meeting the user calculation requirement, so that the energy consumption of the unmannedaerial vehicle in each sub-region is minimum, the energy consumption of the whole unmanned aerial vehicle group is minimum, and the service time of the unmanned aerial vehicle group is prolonged.

The invention belongs to the technical field of ship motion measurement, in particular to a ship turning motion simulation method. The method does not model a propeller, but simulates propeller thrustwith virtual paddle volume force, and simulates ship or submarine rotating motion in water with integral dynamic mesh, so that the simulation precision is improved and the modeling difficulty is simplified, and accordingly designers and testers can quickly and accurately predict the hydrodynamic performance of the ship 's gyration.

The invention provides a power grid line fault diagnosis system and method based on a Hadoop cloud computing system. The system comprises a voltage transformer, a current transformer, a transmitter, a remote terminal, a modem, an optical fiber and a plurality of upper computers. The method includes the steps that voltage and current of a power grid line are acquired in real time and transmitted to the remote terminal; when a power grid has a fault, electrical quantity information of the power grid is acquired and transmitted to a NameNode in a Hadoop cluster and then is assigned to a DateNode to be processed in parallel, and the fault of the power grid is diagnosed and displayed. According to the power grid line fault diagnosis system and method, the electrical quantity information of the power grid line is fully utilized, the current, the wavelet signal rate of the voltage, the wavelet fault rate and the wavelet variation rate are calculated according to the result obtained after wavelet transform reconstruction based on multi-resolution analysis, fusion is performed to form a preliminary credibility distribution function, accurate division is performed after preliminary classification is performed on a fault power grid line set and a non-fault power grid line set, the fault power grid line set is determined finally, and fault power grid lines can be accurately judged out.

The invention discloses a distributed computing method and device, computer equipment and a storage medium. The method comprises the steps of acquiring deployment information of distributed memory grids, and correspondingly setting distributed message queues according to the deployment information; obtaining computing tasks transmitted by message middleware, and executing the computing tasks in corresponding computing nodes of the distributed memory grids to obtain computing results; and obtaining computing logic corresponding to the computing results, and storing the computing results in corresponding to-be-stored nodes in the distributed memory grids. According to the method, the distributed message queues are formed by utilizing local cache addresses of the distributed memory grids; anddata stored in each distributed message queue is sufficient to meet the computing requirements of the current grid, and does not need to be obtained from other distributed message queues, so that thecomputing efficiency of distributed computing is improved.

The invention relates to a method for monitoring and evaluating the operational efficiency grade of a bag type dust remover, in particular to a method for monitoring the working condition of the bag type dust remover in a real-time online manner and for evaluating the operational efficiency grade of the bag type dust remover. The invention aims at solving the problem of providing a system for real-time monitoring of operational efficiency of the bag type dust remover and for quick evaluation of the operational efficiency grade of the bag type dust remover. The adopted scheme is as follows: the system comprises an automatic monitoring device for the operational efficiency grade of the bag type dust remover and a computer system, wherein the automatic monitoring device can realize the automatic real-time measurement of a working current value of a draught fan and the pressure difference value between an inlet and an outlet of the bag type dust remover in working and communicates with the special computer system; through the communication between the computer system and the automatic monitoring device, the setting and the data transfer of the automatic monitoring device are completed; and the quick evaluation of the operational efficiency grade of the bag type dust remover is realized by storing, analyzing and evaluating collected data in real time and outputting evaluation results through the computer system.

The invention relates to the field of wideband identification of ground radar, in particular to a method for extracting a wideband high-resolution range profile based on energy and peak combination. Atarget is preliminarily tracked in a narrow-band mode, and signals are identified in a broadband in the next scanning period. The identification includes an energy summation and maximum peak selection step, an energy and peak combination step, and a target extraction step. Different identification mode weights are adopted in the identification process. Thus, the probability of target extraction is improved.

The invention discloses a real-time risk control variable calculation method based on a distributed streaming calculation engine. The calculation method comprises the steps that a data source configuration module is adopted, wherein a user configures data sources of various calculation dependencies; The variable definition module can be used for developing various variables by using SQL (Structured Query Language); The variable calculation module is used for capturing data from a pre-configured data source according to the calculation request, constructing a memory table, executing all SQL ina related variable pool once, and outputting a calculation result; According to the real-time risk control variable calculation method based on the distributed streaming calculation engine, the variable calculation engine and system researched and developed by the invention are adopted, personnel (such as business personnel and developers) familiar with SQL language can define variables through the system, the variable development period is greatly shortened, and the cost is saved.

The invention discloses a fitness facility exercise data acquisition system based on a distance measurement technology. A fitness facility comprises multiple weights; a hole is formed in each weight;the exercise data acquisition system comprises a bolt, a data acquisition module and a power supply device; when in use, the bolt is inserted into the hole of the weight to be selected; the data acquisition module comprises a shell, and a distance measuring sensor and a control module arranged in the shell; the distance measuring sensor is used for measuring a first distance between the data acquisition module and the bolt when the weight to be selected is located at a static non-suspended state; the control module is used for calculating the total weight of the weights carried by the bolt according to the first distance and a weight-distance relation mapping table; the distance measuring sensor is further used for measuring a second distance between the data acquisition module and the bolt when the weight to be selected is located at a motion state; and the control module calculates real-time exercise journey data of the weight to be selected according to the first distance and the second distance. Therefore, the exercise data of the weights can be accurately measured, demands under different scenarios can be met, and the universality is high.

The invention discloses an AI unloading optimization method for a random task in an industrial Internet of Things, and the method employs an unmanned plane as an edge server, builds an edge calculation entity network of a two-layer unmanned plane, is used for unloading and processing the random task, and optimizes an edge calculation model through an artificial intelligence method. An NIO method is provided for specific task processing and energy consumption when a communication fault occurs in a production area in a dangerous operation environment, and a system unloading model is constructed according to equipment information collected by an unmanned aerial vehicle under an edge computing entity network of two layers of unmanned aerial vehicles. Firstly, an unloading decision of nearby unloading of dense equipment and centralized unloading of dispersed equipment is proposed based on the distance, then energy consumption optimization processing is performed through a Lyapunov optimization method, and finally, an optimal scheme of computing resource allocation and unmanned aerial vehicle deployment is obtained based on a DDPG-G algorithm. According to the method, the problem of task unloading failure caused by communication faults in the dangerous operation process is solved, and system energy consumption optimization is achieved.

The invention discloses a three-dimensional space intelligent positioning and distributed edge computing support method for the Internet of Things, and the method comprises the following steps: acquiring energy consumption of each anchor node and distance information between each anchor node and Internet of Things equipment, and constructing an available anchor node set for each piece of Internet of Things equipment; according to the included angle relation between the anchor nodes, selecting four non-coplanar anchor nodes from the available anchor node set to construct a positioning unit set of the Internet of Things equipment; solving the positioning coordinates of the Internet of Things equipment through the anchor node information in the positioning unit set, and solving the position of the target Internet of Things equipment through the mean value of multiple times of positioning; acquiring positions and task request types of all the Internet of Things devices, dynamically executing path planning by using an unmanned aerial vehicle base station carrying an edge server so as to process requests of the plurality of Internet of Things devices, and determining an unloading decision; enabling the unmanned aerial vehicle to navigate according to the optimal path plan and get close to the Internet of Things equipment, and enabling the Internet of Things equipment to carry out edge calculation unloading.

The invention discloses a data center-based data display method and system. The method comprises the following steps of: creating a data display area; adding components in the data display areas; adding data on the components; realizing data linkage between related components through setting backhaul reference; and finally displaying the data. The system comprises a data display area creation module, a component adding module, a data adding module, a data linkage module and a data dissemination module. Through the method and system, powerful support can be provided for subsequent multilayer data display, manifestation of complex data is more sufficient, convenience is brought to big data analysis, and demands for calculating and monitoring various types of big data in real time can be satisfied.

The invention discloses a reconfigurable pooling operation module structure oriented to a neural network, comprising a dynamic allocation module and a pooling calculation module, wherein the pooling calculation module is used for completing a calculation process of a pooling layer, realizes different pooling methods under the action of the dynamic allocation module and dynamically allocates a calculation scale of a pooling operation; and a current input queue reads from a data storage device after the previous calculation, and a result is transmitted to the data storage device by virtue of anoutput queue. The invention also discloses an implementation method of the reconfigurable pooling operation module structure oriented to the neural network, by virtue of an external control signal, each unit of the dynamic allocation module realizes control on the pooling calculation module, and according to control signal sent out by all the units, the calculation module carries out a corresponding calculation operation and gives an accurate output result. The technical scheme not only can reduce complexity and time of a calculation process and lower power consumption but also improves calculation flexibility and unit utilization rate.

The invention discloses a scheduling method for directed acyclic graph tasks in a heterogeneous multi-core processor system, and belongs to the technical field of computer systems. The method comprises the following steps: acquiring parameters; constructing a priority list of sub-tasks in the DAG; iteratively solving a parameter alpha; and updating the parameter alpha according to a scheduling result obtained by each iteration, terminating the iteration after a judgment condition is met, and outputting a task scheduling result. According to the invention, the task scheduling scheme is iteratively solved by improving the earliest completion time algorithm in the heterogeneous computing environment, the idle period caused by the communication overhead between the processors is reduced, the computing resources of the heterogeneous processors are fully utilized, the completion time of the whole task is shortened, and the real-time requirement of the task is met.