84results about How to "Shorten calculation time" patented technology

The embodiment of the invention discloses an application program push method and a server. The method comprises the steps of acquiring an application program list corresponding to the application program installed in a terminal; calculating the similarity value of the first application program and the second application program according to a pre-stored calculation instruction; detecting whether the similarity value is more than the preset similarity threshold; if so, pushing the second application program corresponding to the similarity value to the terminal. With the adoption of the method, the application program push efficiency can be increased.

The invention provides a task allocation method and device. The method includes the steps that a current computational node is determined as a computation priority type computational node or a storage priority type computational node according to resource information corresponding to each computational node in a distributed computational model; a target computational task is acquired and analyzed into at least two computational processes with the time sequence; the computational processes are analyzed, and the current computational process is analyzed into at least two parallel tasks; each parallel task is analyzed in order to determine attribute information corresponding to the parallel task; the current parallel task is determined as a data-intensive parallel task or a compute-intensive parallel task according to the attribute information corresponding to each parallel task; the compute-intensive parallel task is allocated to the computation priority type computational node, and the data-intensive parallel task is allocated to the storage priority type computational node. By means of the technical scheme, the computational duration of the target computational task can be shortened.

The invention provides a soil saturation hydraulic conductivity measuring and calculating method based on infiltration time characteristic parameters. In the method, the quantitative relation of the characteristic parameters (skewness, kurtosis and variation coefficients) of soil saturation hydraulic conductivity and infiltration time data is built on the basis of analyzing Philip-Dunne infiltration data under three soil texture conditions of clay, loam, sand and the like, and the water content change of soil before and after infiltration does not need any consideration when the method is utilized to measure and calculate the soil saturation hydraulic conductivity, and only the accumulative infiltration time which corresponds to different water levels in a Philip-Dunne infiltration meter needs to be observed, therefore simplifying the calculating process of the saturation hydraulic conductivity and increasing the measuring efficiency.

The invention discloses an aircraft winglet expansion amount optimization method based on a Gaussian pseudo-spectrum method, which comprises the following steps of: establishing constraint conditionsof a winglet expansion track of a climbing section of a near-space variable-wing aircraft, including boundary constraint, path constraint, state variable and control variable constraint; establishinga performance index function of the small wing telescopic track of the climbing section of the near-space variable-wing aircraft, wherein the performance index function is established according to a target with the most fuel-saving effect; discretizing the established continuous optimization control problem by adopting a Gaussian pseudo-spectrum method so as to convert the discretized continuous optimization control problem into a nonlinear programming problem, and solving the nonlinear programming problem to obtain an optimal near-space variable-wing aircraft climbing section winglet telescopic track. According to the method, the expansion amount of the winglet of the climbing section of the near-space variable-wing aircraft can be optimized within a short time, and the control target ofthe minimum fuel consumption is achieved.

The invention discloses an implicit staggered-grid finite difference elastic wave numerical simulation method and device. The method comprises the steps that a dispersion relation expression is obtained based on an implicit staggered-grid finite difference scheme and a plane wave theoretical formula; (M+1) discrete points are evenly taken within the value range of beta in the dispersion relation expression, and the left and right sides of the dispersion relation expression are set to be equal on the discrete points so that a linear algebraic equation set is obtained; the linear algebraic equation set is solved so that the implicit staggered-grid finite difference coefficient am and constant b are obtained; an elastic wave equation is solved based on the implicit staggered-grid finite difference coefficient am, the constant b and the implicit staggered-grid finite difference format; and a wave field diagram is generated according to the solving result and wave field analysis is performed according to the wave field diagram. The solving error of the elastic wave equation can be reduced and the numerical solution accuracy can be enhanced so that a more accurate wave field stimulation record can be obtained. When the same calculation error is constrained, calculation time can be reduced and calculation efficiency can be enhanced.

The invention discloses a distributed power supply-containing power distribution network reconstruction method and system based on a binary differential evolution algorithm of a hybrid strategy, and the method comprises: building a mathematic model of power distribution network reconstruction, and enabling the minimum network loss of a power distribution network to serve as a target function; determining a node type after the distributed power supply is connected into the power distribution network; encoding the feasible solution of the power distribution network by adopting an improved binaryencoding mode, and initializing a population; selecting different mutation strategies by adopting a comparison mode of a current generation optimal individual and a previous generation optimal individual based on weight, and detecting and repairing an invalid solution appearing after mutation operation; adopting a heuristic self-adaptive crossover probability calculation mode, adopting a harmonysearch strategy to optimize crossover operation, and finally completing the crossover operation in a mode of taking a binary string as an individual; and updating a harmony library strategy by utilizing a harmony search algorithm to complete selection operation. Compared with the prior art, the method and the system have the advantages of high robustness, good optimization performance, low convergence algebra and the like.

The invention discloses a training method and device of a load prediction model, a readable medium and electronic equipment, and relates to the field of computers, and the method comprises the steps:obtaining training data features of a plurality of historical load prediction tasks and a selected historical algorithm; taking the training data characteristics of each historical load prediction task and the selected historical algorithm as one piece of training data, and performing machine learning according to each piece of training data to obtain an algorithm selection model; obtaining a plurality of pieces of sample load data of the current load prediction task, and extracting current data characteristics of each piece of sample load data; predicting a current algorithm matched with thecurrent data feature from all historical algorithms through an algorithm selection model; obtaining at least one hyper-parameter corresponding to the current algorithm; and according to the current algorithm, the at least one hyper-parameter and each sample load data, training to obtain a current load prediction model. According to the technical scheme, the calculation time can be shortened, the calculation resources are saved, and the manual intervention degree is reduced.

The invention discloses a social media hot topic extraction method and system. The hot topic extraction method comprises the following steps: filtering and concluding collected Internet text information data, then gathering Internet data information of the same topic by adopting a text clustering algorithm, clustering the Internet data information, and generating a text abstract for the selected topic. The hot topic extraction system comprises a data screening calculation module, a topic clustering module and a topic abstract extraction module, wherein the data screening calculation module isused for filtering and concluding the collected Internet text information data; the topic clustering module adopts the text clustering algorithm to cluster the filtered Internet text information dataof the same topic; and the topic abstract extraction module is used for generating a text abstract from the corresponding clustered Internet text information data according to a topic selected by a user so as to complete hotspot extraction. The method and the system improve the extraction quality of the social media hot topics and have relatively high practical value.

The invention provides a method and device for mapping physical resource of a PDS CH (physical downlink shared channel). Main contents are as follows: a subframe area is divided into various mapping types according to differences of all channel types and signal types of a downlink subframe and / or a special subframe in a wireless frame, when mapped physical resource of PDS CH in a subframe is required to be determined at the aim of the subframe, channel types and signal types capable of being mapped therein are determined according to the mapping types of the subframe, after RE occupied by other channels and signals of the PSD CH are further removed, residual RE is the physical resource mapped by the PSD CH. According to the scheme of the invention, resource indication modes are not required to be considered and the method and the device can be applicable to various systems with different bandwidths, so that the resource mapping ways of PSD CH can be effectively simplified and unified, computing time at the aim of PDSCH resource mapping table is reduced, and further seamless switching in different bandwidth environment can be accomplished quickly.

The invention belongs to the technical field of video encoding and decoding, and particularly relates to a data processing method, a data processing device, a computer readable medium and electronic equipment. The data processing method comprises the following steps: obtaining quantized data obtained by encoding a video image, wherein the quantized data is a numerical matrix comprising a zero value and a non-zero value; obtaining non-zero value position information of the non-zero value in the quantized data, and determining a non-zero region in the quantized data according to the non-zero value position information, the non-zero region being a matrix row or a matrix column in the numerical matrix; performing inverse quantization decoding on the quantized data to obtain inverse quantization data; and performing inverse transformation decoding on the inverse quantization data according to the non-zero region to obtain reconstruction data of the video image. According to the method, thecalculation duration and the calculation capability requirement of inverse transformation encoding can be dynamically reduced according to the distribution condition of the non-zero region in the quantized data, so that the calculation speed can be increased, and calculation resources are saved.

The invention discloses a method for calculating energy flux density distribution of focusing light spots of a tower type solar heat absorber. A heat absorber focusing light spot energy flux density dynamic simulation model of a tower type solar heliostat field is established by utilizing a heat absorber light spot equivalent model. According to the average energy flow density in the conical surface of the projection light of the heliostat, based on coordinate rotation transformation and energy superposition, energy flow density values on grid points of the receiving surface of the heat absorber are obtained through conversion, and then energy flow density distribution of the heat absorber is obtained. By adopting the method, the energy flow density values on all grid points do not need tobe directly calculated, so that the calculation amount is effectively reduced, the calculation duration can be effectively shortened, and the calculation efficiency is improved.

The invention provides a geographic location matching method and device, a storage medium and a server. The method comprises the steps that first geographic location data and second geographic location data are determined, and the positioning accuracy is determined; according to the first geographic location data, the second geographic location data and the positioning accuracy, a first grid wherethe first geographic location data are located and a second grid where the second geographic location data are located are calculated; and the first grid and the second grid are geographically matched. The technical scheme provided by the embodiment of the invention can complete geographical location matching effectively and quickly, improves the geographical location matching efficiency, and shortens the matching time.

The invention provides a method and devices for computing station distance and relates to the field of communication. When the nearest station distance of a to-be-deployed station is computed, computing duration is reduced, and efficiency of computing the nearest station distance is improved. The method comprises the steps of obtaining station information of all stations in a first area, wherein the stations comprise deployed stations and the to-be-deployed station; determining a second area according to the station information of the to-be-deployed station and first preset station distance, wherein the second area is a subarea of the first area; determining the current station as a candidate station if it is determined that the station information of a current station in the first area indicates that a geographic position of the current station is located in the second area; determining candidate distance between the to-be-deployed station and each candidate station; and determining atarget station nearest to the to-be-deployed station and the station distance between the target station and the to-be-deployed station according to each piece of candidate distance.

The invention discloses a data synthesis method and device, data synthesis equipment and a storage medium, and belongs to the technical field of data processing. The method comprises the following steps: acquiring a two-dimensional image of a target object, and performing three-dimensional mapping on the two-dimensional image of the target object to obtain a three-dimensional model of the target object; obtaining a three-dimensional model of the additional data; and matching the three-dimensional model of the target object with the three-dimensional model of the additional data through the texture coordinates of the target pixels, and obtaining an image of the target object for synthesizing the additional data according to a matching result. A three-dimensional model of a target object isobtained by performing three-dimensional mapping on a two-dimensional image of the target object, and the three-dimensional model of the target object and the three-dimensional model of the additionaldata are matched in a texture coordinate mode to obtain an image of the target object for synthesizing the additional data, so that three-dimensional data synthesis can be realized through a 2D camera, and the cost is relatively low. In addition, multiple modeling is not needed, so that the calculation time is saved, and the data synthesis efficiency is higher.

A carrier frequency offset estimation method of baseband signal and device thereof, the carrier frequency offset estimation method comprises the steps of baseband signal acquisition, complex signal calculation, complex signal frequency spectrum shifting, complex signal resampling, useful spectrum line frequency point sequence number acquisition and carrier frequency offset estimation, etc. As thecomplex signal after spectrum shifting is resampled, the sampling rate of the system can be matched to a fixed value, so that FFT processing with a fixed number of points can be adopted to obtain a spectral sequence including useful spectral lines, so that the system can obtain the frequency point number corresponding to the useful spectral line without increasing the number of FFT points, but also simplifies the calculation process of obtaining the frequency point serial number. On the premise of guaranteeing the higher precision of carrier frequency offset estimation, it can speed up the processing speed, further reduce the calculation time and the running cost of the system, and finally help to improve the real-time performance of OQPSK signal in demodulation processing and carrier frequency offset estimation processing.

The invention discloses an image restoration method based on overlapping combination sparse high-order total variation. The method comprises the following steps: S30, establishing an image restorationhybrid model of an original image; S40, obtaining a constrained optimization problem; S50, splitting the constrained optimization problem into independent sub-problems to obtain an augmented Lagrangefunction; S60, solving each variable quantum problem of the augmented Lagrange function by adopting a Lagrange multiplier in the kth iteration process to obtain a target function in the kth iterationprocess, and further obtaining a restored image, a first variable parameter and a second variable parameter in the kth iteration process; S70, updating the Lagrange multiplier according to the restored image, the first variable parameter and the second variable parameter to obtain a Lagrange multiplier in the next iteration process; and S80, judging whether the stop parameter in the kth iterationprocess is less than or equal to a stop threshold, if yes, determining a restored image, and if not, setting k to be equal to k + 1, and returning to execute the step S60.

The method is suitable for the technical field of image processing. The invention provides a depth information acquisition method, a depth information acquisition system and terminal equipment. Two images in the same scene are acquired through double cameras, the matching cost of each pixel point in the two images is calculated respectively, and cost aggregation in a preset number of directions isperformed on each pixel point according to the matching cost of each pixel point to obtain a disparity map of the two images, so that the calculation time can be effectively shortened, and the powerconsumption can be reduced; and after the disparity maps of the two images are converted into the depth maps, the depth maps are respectively mapped back to the two images, and the depth information can be directly applied to the original image.

The invention discloses a simulation calculation method for the air inlet amount of a cabin heat exchanger under a transient driving condition, which comprises the following steps of: determining sampling points ai, performing steady-state simulation calculation on each sampling point ai, and outputting a statistical average value Q1 of the simulated air inlet rate of each sampling point ai; fitting the sampling points ai with the corresponding average value Q1 of the air inlet volume to obtain a relational expression between the vehicle speed and the air inlet volume, selecting a test point bi, and substituting the test point bi into the relational expression to obtain a statistical average value Q2 of the calculated air inlet volume of the test point; carrying out steady-state simulation calculation on the test point bi, and obtaining the statistical average value Q2b of the simulation air inlet amount of the test point; at the moment, determining a precise relational expression; at the moment, re-planning the sampling point ai; importing the vehicle speed under the transient driving working condition into the determined accurate relational expression, and obtaining a relation curve of the air inlet amount of the cabin heat exchanger under the transient working condition, wherein the air inlet amount of the cabin heat exchanger comprises the air inlet quantity of a radiator and the air inlet quantity of a condenser; predicting transient driving conditions by discretizing a transient process into a plurality of steady-state calculation processes.

The invention discloses a data processing method and device. The data processing method comprises the following steps: acquiring a data distribution algorithm of a distributed system for achieving load balancing based on particle swarm optimization; according to a data distribution algorithm for achieving load balancing based on particle swarm optimization, calculating an optimal position of a particle for load balancing in the distributed system; obtaining a data distribution algorithm for achieving optimal storage based on particle swarm optimization of the distributed system; and optimizingthe storage space required by each node in the distributed system according to a data distribution algorithm for realizing optimal storage based on particle swarm optimization and the optimal particle position of load balancing. According to the method and device, the technical problems of long calculation time and low efficiency caused by the fact that the existing full comparison calculation research adopts a branch and bound method to complete data distribution of full comparison calculation are solved.

The invention discloses an improved continuum structure topological optimization post-processing method. The improved continuum structure topological optimization post-processing method is characterized by adopting a large grid size, namely, the size of a finite element analysis unit being larger than a default value 1, and when topological optimization is conducted on a target model, effectivelyavoiding a fine columnar structure caused by too many grids; and performing binarization processing on the optimization model, analyzing the optimization model, extracting a zigzag boundary line, obtaining a target discrete angular point set through numerical calculation, performing curve fitting by taking the target discrete angular point set as a sample point, performing further interpolation processing on the obtained new boundary line to generate a topological structure with a smooth boundary, and performing entity processing on the topological structure. According to the method provided by the invention, the sawtooth boundary of the optimized structure is effectively eliminated, the original design requirement of the optimized structure is ensured, the calculation difficulty is reduced, the calculation duration is shortened, the model manufacturability is enhanced, and the difficulty of a topological optimization post-processing procedure is effectively reduced.

The distributed score card model establishing method comprises the following steps: binning sample data of each label variable into N sample intervals, and calculating chi-square values of adjacent intervals; combining the adjacent intervals corresponding to the minimum chi-square value into a new interval, judging whether the number of the combined intervals reaches the target number or not, if so, executing S3, or otherwise, executing S4; S3, carrying out monotonicity testing on the chi-square values, calculated by the combined intervals, of the adjacent intervals, if the chi-square values are monotonous, executing the step S5, and if not, combining the adjacent intervals corresponding to the minimum chi-square value into a new interval, and executing the step S3; S4, recalculating the chi-square value of each combined adjacent interval and executing the step S2; and S5, judging whether the proportion of the sample data of the interval exceeds a set proportion value, if so, abandoning the label variable, otherwise, calculating an information credit investigation value of the interval, selecting the label variable of which the information credit investigation value reaches a preset requirement as a target label variable, and substituting the sample data of the target label variable into model training.

The embodiment of the invention discloses a neural network training method and device, equipment and a computer readable storage medium. The method comprises the following steps: converting parameters of a preset network layer in a to-be-trained model from a first data type to a second data type to obtain a prediction model; performing forward propagation on the obtained training sample in the prediction model until the training sample is transmitted to an output layer of which the parameter type is the first data type, and obtaining a prediction value of the first data type; according to the difference between the predicted value of the first data type and the true value of the training sample, performing parameter updating on the to-be-trained model to obtain an updated model; continuously updating the parameters of the updated model until a training end condition is met, and generating a target model; the target model is used for determining a plurality of prediction labels corresponding to a plurality of label dimensions for an input prediction sample. According to the invention, the efficiency of neural network training can be improved.

The invention relates to a hybrid optimization method and system for auxiliary edge calculation of an unmanned aerial vehicle, and the method comprises that: taking the minimum energy consumption required by an auxiliary edge calculation model of the unmanned aerial vehicle as a target fitness function; performing iterative optimization solution on a target fitness function by utilizing a probability selection strategy of a hybrid artificial bee colony algorithm taking an imitated flower pollination algorithm framework as a reference, and judging whether a premature convergence phenomenon exists based on individual fitness value change and population diversity, so that the global optimization capability of the hybrid optimization algorithm is stronger, and then optimal calculation task unloading, calculation resource allocation and unmanned aerial vehicle deployment position schemes are realized. The method is high in calculation speed, high in calculation precision and suitable for solving the edge calculation service problem of large-scale mobile user equipment calculation unloading and task allocation.

Disclosed herein is a method of storing or transferring an image along with an image-processing parameter for processing the image. When storing or transferring a second image for storage or transfer which has a second resolution for storage or transfer differing from a reference resolution, a reference image-processing parameter set according to a reference resolution image having the reference resolution is corrected based on a difference between the reference resolution and the second resolution, so that it becomes a second parameter corresponding to the second image. The second parameter obtained by the correction is stored or transferred along with the second image.

The invention discloses a method and system for identifying a relationship between a user meter and a meter box, equipment and a storage medium; the method comprises the steps: firstly, calculating voltage correlation based on a Pearson's correlation coefficient algorithm to rapidly and accurately judge the attribution relationship between the user meter and a branch, greatly shortening the calculation time, and reducing the data size of subsequent calculation; and then constructing a multiple linear regression model, and accurately obtaining the phase of the single-phase user meter based on a mean square error calculation result of each model; finally, constructing combination types of relations between all household meters and meter boxes under each branch, numbering the combination types in sequence, calculating to obtain offset loss of each meter box under the combination types, and performing principal component analysis, thereby quickly and accurately identifying the relation between the household meters and the meter boxes under each branch. Besides, the electricity consumption is adopted as analysis data, and the current I, the voltage U and the active power P are adopted, so that the calculation result is not interfered by empty households, and the accuracy of the identification result is further ensured.

The invention provides a bayesian optimization design method of a metal-dielectric multilayer columnar optical nano antenna, which comprises the following steps: S1, constructing a structure parametercandidate set according to antenna design requirements; s2, determining an optimization objective function as a far-field directivity coefficient, and setting a constraint function at the same time;s3, setting an initial training parameter set and obtaining a corresponding optimization objective function value and a corresponding constraint function value, taking the three sets as a training set, and setting an initial value of iteration times k; s4, training hyper-parameters in the Gaussian model according to the training set; s5, selecting a next candidate parameter; s6, calculating an objective function value and a constraint function value corresponding to the candidate parameter; and S7, judging whether the objective function values corresponding to the candidate parameters meetingthe constraint conditions meet cut-off conditions or not. The method has the advantages that the method is optimized through Bayesian optimization, the calculation time can be effectively shortened, the design efficiency is improved, and the purpose of quickly optimizing the antenna scattering efficiency is achieved.

The invention discloses a cluster calling system, method and device, and the system comprises a data receiving and verification node which receives data and carries out the verification and initialization of data, and obtains target data; a database updating node used for receiving the target data, establishing a current database according to the target data, and sending the current database tothe logic computing node, the diagnosis node and the standby database updating node in the cluster; a load balancing node used for carrying out reverse proxy on the received calculation request and shunting the calculation request to the logic calculation node; a logic computing node used for receiving the computing request and operating the corresponding computing engine to compute; a main noderesponsible for storing node information in a cluster and switching versions to coordinate the work of each node; anda diagnosis node which supports operation and maintenance personnel to diagnose thedata in the current database and the historical database. The slave nodes are subdivided, so that the calculation duration of a single request is shortened, and the running speed is increased.

The invention provides a narrowband Internet of Things time-frequency offset estimation method and device, and a storage medium. The method comprises the steps that at least one piece of preset sampling rate subframe data is acquired; performing self-correlation operation on a plurality of orthogonal frequency division multiplexing symbols in the preset sampling rate subframe data to generate a plurality of self-correlation operation values; when the plurality of self-correlation operation values meet a preset time offset estimation condition, determining a time offset value of a main synchronization signal of the narrowband Internet of Things; carrying out frequency offset value calculation on a plurality of orthogonal frequency division multiplexing symbols in the preset sampling rate subframe data by adopting a frequency offset estimation algorithm, and generating a frequency offset value of a main synchronization signal of the narrowband Internet of Things; and outputting the time offset value and the frequency offset value of the main synchronization signal of the narrowband Internet of Things. In this way, the calculation amount of the time-frequency offset of the narrowband Internet of Things is reduced, and then the calculation duration is shortened.